diff --git a/.gitattributes b/.gitattributes index a6344aac8c09253b3b630fb776ae94478aa0275b..52373fe24473b1aa44333d318f578ae6bf04b49b 100644 --- a/.gitattributes +++ b/.gitattributes @@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text *.zip filter=lfs diff=lfs merge=lfs -text *.zst filter=lfs diff=lfs merge=lfs -text *tfevents* filter=lfs diff=lfs merge=lfs -text +tokenizer.json filter=lfs diff=lfs merge=lfs -text diff --git a/mlc-chat-config.json b/mlc-chat-config.json new file mode 100644 index 0000000000000000000000000000000000000000..79d493e7a2789a7b8baf86485f83e8649131fda7 --- /dev/null +++ b/mlc-chat-config.json @@ -0,0 +1,79 @@ +{ + "version": "0.1.0", + "model_type": "gemma", + "quantization": "q0f16", + "model_config": { + "hidden_size": 3072, + "hidden_activation": "gelu_pytorch_tanh", + "intermediate_size": 24576, + "attention_bias": false, + "num_attention_heads": 16, + "num_key_value_heads": 16, + "head_dim": 256, + "num_hidden_layers": 28, + "rms_norm_eps": 1e-06, + "vocab_size": 256000, + "position_embedding_base": 10000.0, + "context_window_size": 8192, + "prefill_chunk_size": 2048, + "tensor_parallel_shards": 1, + "max_batch_size": 80 + }, + "vocab_size": 256000, + "context_window_size": 8192, + "sliding_window_size": -1, + "prefill_chunk_size": 2048, + "attention_sink_size": -1, + "tensor_parallel_shards": 1, + "temperature": 1.0, + "presence_penalty": 0.0, + "frequency_penalty": 0.0, + "repetition_penalty": 1.0, + "top_p": 1.0, + "tokenizer_files": [ + "tokenizer.model", + "tokenizer.json", + "tokenizer_config.json" + ], + "tokenizer_info": { + "token_postproc_method": "byte_fallback", + "prepend_space_in_encode": false, + "strip_space_in_decode": false + }, + "conv_template": { + "name": "gemma_instruction", + "system_template": "{system_message}", + "system_message": "", + "system_prefix_token_ids": [ + 2 + ], + "add_role_after_system_message": true, + "roles": { + "user": "user", + "assistant": "model" + }, + "role_templates": { + "user": "{user_message}", + "assistant": "{assistant_message}", + "tool": "{tool_message}" + }, + "messages": [], + "seps": [ + "\n" + ], + "role_content_sep": "\n", + "role_empty_sep": "\n", + "stop_str": [ + "" + ], + "stop_token_ids": [ + 1, + 107 + ], + "function_string": "", + "use_function_calling": false + }, + "pad_token_id": 0, + "bos_token_id": 2, + "eos_token_id": 1 +} \ No newline at end of file diff --git a/ndarray-cache.json b/ndarray-cache.json new file mode 100644 index 0000000000000000000000000000000000000000..be0fe26ed33f9ac7f1e5fceba5c4b1fb7f4522c2 --- /dev/null +++ b/ndarray-cache.json @@ -0,0 +1,2726 @@ +{ + "metadata": { + "ParamSize": 170, + "ParamBytes": 17075361792.0, + "BitsPerParam": 16.0 + }, + "records": [ + { + "dataPath": "params_shard_0.bin", + "format": "raw-shard", + "nbytes": 1572864000, + "records": [ + { + "name": "model.embed_tokens.weight", + "shape": [ + 256000, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 1572864000, + "byteOffset": 0 + } + ], + "md5sum": "60ae32290a73af4d897daca869ced218" + }, + { + "dataPath": "params_shard_1.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.0.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "bbe91c0e9a7cc6b67a2ac8e433ad38ca" + }, + { + "dataPath": "params_shard_2.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.0.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "d6084ff7e8369ae1a3004ce30a5a3a8b" + }, + { + "dataPath": "params_shard_3.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.0.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "180aa61e6f9c9532539c4257fbb58992" + }, + { + "dataPath": "params_shard_4.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.1.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "f4addb4fe0430f3cdb3d2319bb44ba62" + }, + { + "dataPath": "params_shard_5.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.1.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "b7691ec6a77ca8625c98f97bd1635b06" + }, + { + "dataPath": "params_shard_6.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.1.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "160e7c76a13c050aeec492a2e2d72aab" + }, + { + "dataPath": "params_shard_7.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.1.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "b1d4c02924ac0bac0c95681ffe5d7163" + }, + { + "dataPath": "params_shard_8.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.2.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "c49146f24474b26a7832f3ab4a83d39a" + }, + { + "dataPath": "params_shard_9.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.2.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "f01062ca986324737db0f17aaf5b99f2" + }, + { + "dataPath": "params_shard_10.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.2.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "484e852cde8ad23c2b0d8e5cfa5b5738" + }, + { + "dataPath": "params_shard_11.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.2.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "4a09c9ce10ad9282efe2ac2b82221647" + }, + { + "dataPath": "params_shard_12.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.3.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "96f11c66cf3c5ca2e4a8521761f98e23" + }, + { + "dataPath": "params_shard_13.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.3.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "340727bd83ea2235d0074ec50233e934" + }, + { + "dataPath": "params_shard_14.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.3.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "5c8c43c7fafd195e9d744b9ef2fe7217" + }, + { + "dataPath": "params_shard_15.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.3.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "1777bcd65b8d2726f1003d2630ce47c9" + }, + { + "dataPath": "params_shard_16.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.4.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "9bb0bb3b2b4befa28a13d43b5c2ba0e8" + }, + { + "dataPath": "params_shard_17.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.4.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "d75e37069c46ecdd74b80f8cc06cb6d1" + }, + { + "dataPath": "params_shard_18.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.4.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "f61a8e4f3be662e511bf8d2cfa546a74" + }, + { + "dataPath": "params_shard_19.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.4.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "fd2ad1fb42458ff412cf6ab0eacd07a9" + }, + { + "dataPath": "params_shard_20.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.5.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "a4a0bec88d38218586cc0fecfe96f4a5" + }, + { + "dataPath": "params_shard_21.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.5.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "abd666001603d179c6fe9fef7ca52b81" + }, + { + "dataPath": "params_shard_22.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.5.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "40daa99215dbc17975c3167e09ad31e5" + }, + { + "dataPath": "params_shard_23.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.5.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "71be785d02736b5922a6cae0c8d1a831" + }, + { + "dataPath": "params_shard_24.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.6.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "fb2f730b5d66c9cdbc798358bef79ebe" + }, + { + "dataPath": "params_shard_25.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.6.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "5165b9b2823cb5bd00219adede68d307" + }, + { + "dataPath": "params_shard_26.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.10.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "0f3ad4fb9d4b1c11625a92ab26b15565" + }, + { + "dataPath": "params_shard_27.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.10.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "8857080ba55c0ca02c1bfbd0204a05f5" + }, + { + "dataPath": "params_shard_28.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.10.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "b00a53f899e45d7276e99354143f1c61" + }, + { + "dataPath": "params_shard_29.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.10.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "c310bc2010941c90ab5501059839d251" + }, + { + "dataPath": "params_shard_30.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.11.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "fb0a228dbb72498a441e98248a05f195" + }, + { + "dataPath": "params_shard_31.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.11.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "1bca1e158bbac017220d40d0d261992d" + }, + { + "dataPath": "params_shard_32.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.11.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "a35cff8ebbb396c6b83d2a2259dd72c4" + }, + { + "dataPath": "params_shard_33.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.11.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "a6e50ab55df754b9034fa46797bacb4f" + }, + { + "dataPath": "params_shard_34.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.12.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "f0572dda1a68d882525bcf76be6ca777" + }, + { + "dataPath": "params_shard_35.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.12.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "291457429b0e015a4bcf8807b7fe55ed" + }, + { + "dataPath": "params_shard_36.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.12.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "df565ab47cc6f4afd0f77c7f05631b6c" + }, + { + "dataPath": "params_shard_37.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.12.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "978f92f008c4dd62822eef690eade5ea" + }, + { + "dataPath": "params_shard_38.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.13.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "f3d368b4cab5b1fc97723ec6a1821f77" + }, + { + "dataPath": "params_shard_39.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.13.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "898fa212a2a0c5d7dd6c6e8db2e1a71d" + }, + { + "dataPath": "params_shard_40.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.13.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "86fca1e31ddb2efa2ffd3876e106c10a" + }, + { + "dataPath": "params_shard_41.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.13.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "da402edecdd2e1a3c251ee0610a536e0" + }, + { + "dataPath": "params_shard_42.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.14.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "86a91af650313b154a76a83215b76df8" + }, + { + "dataPath": "params_shard_43.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.14.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "126bb57f096bf441e7f88808078630ad" + }, + { + "dataPath": "params_shard_44.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.14.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "87fbd15936e44077592e028134bc0376" + }, + { + "dataPath": "params_shard_45.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.14.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "00f38ed56e2d90c0227acd3acdcf0241" + }, + { + "dataPath": "params_shard_46.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.15.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "bbad0dd8e2e23dda60e4e061454e3abd" + }, + { + "dataPath": "params_shard_47.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.15.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "09118299bdefbae7d50ffa91c5b224d1" + }, + { + "dataPath": "params_shard_48.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.6.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "053d8c972995b8d83e61744e6e314c7f" + }, + { + "dataPath": "params_shard_49.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.6.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "ca72202cff818ef742ac296bc2600926" + }, + { + "dataPath": "params_shard_50.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.7.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "3686942021d12436094f0d9b3ef7b837" + }, + { + "dataPath": "params_shard_51.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.7.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "8470ce2af9e62816b6f73efaa79bd720" + }, + { + "dataPath": "params_shard_52.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.7.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "0c344d1f3c747dbad5db78df451ccf23" + }, + { + "dataPath": "params_shard_53.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.7.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "54b10013400455dd7cf82060c38bc4be" + }, + { + "dataPath": "params_shard_54.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.8.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "192a3d2fcf7d3c70d3dbc0a93737ca88" + }, + { + "dataPath": "params_shard_55.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.8.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "96d24ad55a29534d380ce2fd2e6c992c" + }, + { + "dataPath": "params_shard_56.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.8.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "2bd0598739a7587684fceeeb39086c6a" + }, + { + "dataPath": "params_shard_57.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.8.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "a1dd21a5a45c4d270e8df09a1b934817" + }, + { + "dataPath": "params_shard_58.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.9.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "29670e1879eaa9af540508e6b3a38f2e" + }, + { + "dataPath": "params_shard_59.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.9.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "4d0e8592ac518ce1aea4e0a1bae33cfd" + }, + { + "dataPath": "params_shard_60.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.9.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "bd91da9cd9fd37fef6c0de05026c2034" + }, + { + "dataPath": "params_shard_61.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.9.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "12bc8d3ab3b5055f5f431c0971b91a78" + }, + { + "dataPath": "params_shard_62.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.15.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "520345638bf42df4cb55d8479ff770fe" + }, + { + "dataPath": "params_shard_63.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.15.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "ac934115b39eeb7519802192e1e103b9" + }, + { + "dataPath": "params_shard_64.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.16.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "63345fb1ae9f9b058b5883f41c33db14" + }, + { + "dataPath": "params_shard_65.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.16.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "dd8d733ee27a5e5dc6d41668d4d7cd6c" + }, + { + "dataPath": "params_shard_66.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.16.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "f5654529a63557c8e4005374d6cf296d" + }, + { + "dataPath": "params_shard_67.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.16.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "3f4abeb620ef7ec88e4b7849c5658975" + }, + { + "dataPath": "params_shard_68.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.17.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "bd9921891a8e96b69c9a169e0b19ad88" + }, + { + "dataPath": "params_shard_69.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.17.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "0c0627ce937b0edd3e98b0dc2b163c5a" + }, + { + "dataPath": "params_shard_70.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.17.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "6a6beccbe8824ac3abb1d1ce498ceb9c" + }, + { + "dataPath": "params_shard_71.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.17.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "f50c090dd54c9b05b0057e1b7afcdfd1" + }, + { + "dataPath": "params_shard_72.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.18.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "f0725fb767165df3f963fce9f3db89d0" + }, + { + "dataPath": "params_shard_73.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.18.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "873b23b9460ae71eddf36a532499c553" + }, + { + "dataPath": "params_shard_74.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.18.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "2e49ca76bb62f6fe4a00a5c8ee53dae7" + }, + { + "dataPath": "params_shard_75.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.18.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "c5af0bf1da75e442102bb2ab9e2b5fdb" + }, + { + "dataPath": "params_shard_76.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.19.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "09ec40932166e5700f99f736b1576724" + }, + { + "dataPath": "params_shard_77.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.19.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "aa670b96411b8d184f4545b37acaa097" + }, + { + "dataPath": "params_shard_78.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.19.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "760fcb364bc239c2e77c39b96a9a7466" + }, + { + "dataPath": "params_shard_79.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.19.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "977188d5ec51682ebc6f673487072a99" + }, + { + "dataPath": "params_shard_80.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.20.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "90770a8cde2bea125b1cd9b51e333131" + }, + { + "dataPath": "params_shard_81.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.20.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "69db499e0e83a0aeb51c212848da685f" + }, + { + "dataPath": "params_shard_82.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.20.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "3c3bd1e3cc7b60660caf1d364538073e" + }, + { + "dataPath": "params_shard_83.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.20.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "14e408e470361728f075ca6e96d812ad" + }, + { + "dataPath": "params_shard_84.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.21.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "91252eb5ac7e8ff526b6a6539373e5ee" + }, + { + "dataPath": "params_shard_85.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.21.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "385639a8f299bf0e1086cf111c79feca" + }, + { + "dataPath": "params_shard_86.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.21.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "e25ade1d0ba5a8f660c1ff40aa876312" + }, + { + "dataPath": "params_shard_87.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.21.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "93afa47a95a2a34d6209880dfdcf2713" + }, + { + "dataPath": "params_shard_88.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.22.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "20d81e0b42f15f8cbe66b37be4ce7a62" + }, + { + "dataPath": "params_shard_89.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.22.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "d3f4a3ac0c3270e6a18d9c69cc8bbc19" + }, + { + "dataPath": "params_shard_90.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.22.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "f41aa07eb326a16bd1d4f25ca61372f3" + }, + { + "dataPath": "params_shard_91.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.22.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "6757f617b416676c1ac978082017a06f" + }, + { + "dataPath": "params_shard_92.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.23.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "1daf7b3195a362019245b4aaac97afa3" + }, + { + "dataPath": "params_shard_93.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.23.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "dd1649b386d47e14b297af0eef72d212" + }, + { + "dataPath": "params_shard_94.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.23.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "e2c261a370277bf7345af16ae56ce3dd" + }, + { + "dataPath": "params_shard_95.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.23.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "d0237722ad61da671a535e24681ac224" + }, + { + "dataPath": "params_shard_96.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.24.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "17032abea04fc43daf6864142a4fd24c" + }, + { + "dataPath": "params_shard_97.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.24.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "2c661f393440ce8b8e3089da0b74df90" + }, + { + "dataPath": "params_shard_98.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.24.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "ae565af5cc648c6d29d43cf53253ea8b" + }, + { + "dataPath": "params_shard_99.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.24.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "2fb95a887ab3b66c052a264e3fe7711f" + }, + { + "dataPath": "params_shard_100.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.25.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "918eb03c29d4fccaa534f6396161eb69" + }, + { + "dataPath": "params_shard_101.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.25.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "9c2be4c01241aca2b838e678f6a4602e" + }, + { + "dataPath": "params_shard_102.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.25.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "5b05018af280f877ac02b185c02a8335" + }, + { + "dataPath": "params_shard_103.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.25.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "b60134b8d6353407694018e6561c309e" + }, + { + "dataPath": "params_shard_104.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.26.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "1e725304f8124408742d3488dfd09f3b" + }, + { + "dataPath": "params_shard_105.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.26.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "ca7dd5e249a888b895aa3e19e01cf5ae" + }, + { + "dataPath": "params_shard_106.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.26.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "3b58c08fd5f34ba34ffb26f67561fab7" + }, + { + "dataPath": "params_shard_107.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.26.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "0236d2e8b4ca91382c411d3618a26be8" + }, + { + "dataPath": "params_shard_108.bin", + "format": "raw-shard", + "nbytes": 150994944, + "records": [ + { + "name": "model.layers.27.mlp.down_proj.weight", + "shape": [ + 3072, + 24576 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 150994944, + "byteOffset": 0 + } + ], + "md5sum": "86a1cb3d42ed8c7cb9c91a923ed048ec" + }, + { + "dataPath": "params_shard_109.bin", + "format": "raw-shard", + "nbytes": 301989888, + "records": [ + { + "name": "model.layers.27.mlp.gate_up_proj.weight", + "shape": [ + 49152, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 301989888, + "byteOffset": 0 + } + ], + "md5sum": "7f26d02cec6df2d416595c0227777d67" + }, + { + "dataPath": "params_shard_110.bin", + "format": "raw-shard", + "nbytes": 75497472, + "records": [ + { + "name": "model.layers.27.self_attn.qkv_proj.weight", + "shape": [ + 12288, + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 75497472, + "byteOffset": 0 + } + ], + "md5sum": "831436910bc5f74746081ba7de504bb9" + }, + { + "dataPath": "params_shard_111.bin", + "format": "raw-shard", + "nbytes": 25165824, + "records": [ + { + "name": "model.layers.27.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 0 + } + ], + "md5sum": "ab01465acddcdcbfc3f42866c1178acb" + }, + { + "dataPath": "params_shard_112.bin", + "format": "raw-shard", + "nbytes": 25516032, + "records": [ + { + "name": "model.layers.0.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 0 + }, + { + "name": "model.layers.0.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 6144 + }, + { + "name": "model.layers.0.self_attn.o_proj.weight", + "shape": [ + 3072, + 4096 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 25165824, + "byteOffset": 12288 + }, + { + "name": "model.layers.1.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25178112 + }, + { + "name": "model.layers.1.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25184256 + }, + { + "name": "model.layers.2.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25190400 + }, + { + "name": "model.layers.2.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25196544 + }, + { + "name": "model.layers.3.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25202688 + }, + { + "name": "model.layers.3.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25208832 + }, + { + "name": "model.layers.4.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25214976 + }, + { + "name": "model.layers.4.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25221120 + }, + { + "name": "model.layers.5.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25227264 + }, + { + "name": "model.layers.5.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25233408 + }, + { + "name": "model.layers.10.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25239552 + }, + { + "name": "model.layers.10.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25245696 + }, + { + "name": "model.layers.11.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25251840 + }, + { + "name": "model.layers.11.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25257984 + }, + { + "name": "model.layers.12.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25264128 + }, + { + "name": "model.layers.12.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25270272 + }, + { + "name": "model.layers.13.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25276416 + }, + { + "name": "model.layers.13.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25282560 + }, + { + "name": "model.layers.14.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25288704 + }, + { + "name": "model.layers.14.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25294848 + }, + { + "name": "model.layers.6.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25300992 + }, + { + "name": "model.layers.6.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25307136 + }, + { + "name": "model.layers.7.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25313280 + }, + { + "name": "model.layers.7.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25319424 + }, + { + "name": "model.layers.8.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25325568 + }, + { + "name": "model.layers.8.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25331712 + }, + { + "name": "model.layers.9.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25337856 + }, + { + "name": "model.layers.9.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25344000 + }, + { + "name": "model.layers.15.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25350144 + }, + { + "name": "model.layers.15.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25356288 + }, + { + "name": "model.layers.16.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25362432 + }, + { + "name": "model.layers.16.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25368576 + }, + { + "name": "model.layers.17.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25374720 + }, + { + "name": "model.layers.17.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25380864 + }, + { + "name": "model.layers.18.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25387008 + }, + { + "name": "model.layers.18.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25393152 + }, + { + "name": "model.layers.19.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25399296 + }, + { + "name": "model.layers.19.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25405440 + }, + { + "name": "model.layers.20.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25411584 + }, + { + "name": "model.layers.20.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25417728 + }, + { + "name": "model.layers.21.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25423872 + }, + { + "name": "model.layers.21.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25430016 + }, + { + "name": "model.layers.22.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25436160 + }, + { + "name": "model.layers.22.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25442304 + }, + { + "name": "model.layers.23.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25448448 + }, + { + "name": "model.layers.23.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25454592 + }, + { + "name": "model.layers.24.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25460736 + }, + { + "name": "model.layers.24.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25466880 + }, + { + "name": "model.layers.25.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25473024 + }, + { + "name": "model.layers.25.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25479168 + }, + { + "name": "model.layers.26.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25485312 + }, + { + "name": "model.layers.26.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25491456 + }, + { + "name": "model.layers.27.input_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25497600 + }, + { + "name": "model.layers.27.post_attention_layernorm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25503744 + }, + { + "name": "model.norm.weight", + "shape": [ + 3072 + ], + "dtype": "float16", + "format": "f32-to-bf16", + "nbytes": 6144, + "byteOffset": 25509888 + } + ], + "md5sum": "f3ad81e901d1821a5626ed9c99f5e4a1" + } + ] +} \ No newline at end of file diff --git a/params_shard_0.bin b/params_shard_0.bin new file mode 100644 index 0000000000000000000000000000000000000000..ca1fd3e199ee32686c2a458488794d7fceb58c3e --- /dev/null +++ b/params_shard_0.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0cbebcb441b6b72286264b79479f70decc2de2fc3228542be84159e6cd838e05 +size 1572864000 diff --git a/params_shard_1.bin b/params_shard_1.bin new file mode 100644 index 0000000000000000000000000000000000000000..6a48dfa37401847afe4fcae55b24c99140602e8a --- /dev/null +++ b/params_shard_1.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ea372b196ac040951076692fa883a590548f52217f472e204fdbc0aa9490de58 +size 150994944 diff --git a/params_shard_10.bin b/params_shard_10.bin new file mode 100644 index 0000000000000000000000000000000000000000..ca786ddb9394592413092eb7c8b9fa450f5f3c68 --- /dev/null +++ b/params_shard_10.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:edfd11bc111b52af23b416137b8797c7210e675327d88461897842da3887a5c3 +size 75497472 diff --git a/params_shard_100.bin b/params_shard_100.bin new file mode 100644 index 0000000000000000000000000000000000000000..dc3ad3b28711b9498955307ee6f602de1b3746dd --- /dev/null +++ b/params_shard_100.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aa81bce7b80f1211cedf4ea95d1636c1ff2640b1b8029aa99ea6bda3d3f96e3a +size 150994944 diff --git a/params_shard_101.bin b/params_shard_101.bin new file mode 100644 index 0000000000000000000000000000000000000000..bcf9ae3f8b4507bbbd48d31ed40af208e12c584d --- /dev/null +++ b/params_shard_101.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1331091b19fa3791520c8e21b081cb6b9dd1957ac62df9fb964e12e1ed28ccad +size 301989888 diff --git a/params_shard_102.bin b/params_shard_102.bin new file mode 100644 index 0000000000000000000000000000000000000000..59b031f396f70498c502aade2ab2009b5abc5a2d --- /dev/null +++ b/params_shard_102.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:301022b938e8673e244ab47e1b2ba493a5bc60e6b1c671c0754ed39ccef38695 +size 75497472 diff --git a/params_shard_103.bin b/params_shard_103.bin new file mode 100644 index 0000000000000000000000000000000000000000..bb9567262120d7d1a93cd36cd967d3aa649731f1 --- /dev/null +++ b/params_shard_103.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a9b0b74e0c1059dbf84bf4e4b6363ef79ba38e1f40493d2cda62938bdbf29809 +size 25165824 diff --git a/params_shard_104.bin b/params_shard_104.bin new file mode 100644 index 0000000000000000000000000000000000000000..9a6e0c0d09affc87edb587c3663347f5a69db19f --- /dev/null +++ b/params_shard_104.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f5a16912f4597de137be9d85645bb01b2d3d08dfda30cbeb7189a83fbc6cab3d +size 150994944 diff --git a/params_shard_105.bin b/params_shard_105.bin new file mode 100644 index 0000000000000000000000000000000000000000..ff30fc682db99406031b95baced1cb9f72afd91e --- /dev/null +++ b/params_shard_105.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:be0e80cf51b82121d27a5aeb65d788160bff981a24fa63dcd3a0927077e54f40 +size 301989888 diff --git a/params_shard_106.bin b/params_shard_106.bin new file mode 100644 index 0000000000000000000000000000000000000000..778abc8062fb005934067d119eb525f72ee3f3b2 --- /dev/null +++ b/params_shard_106.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:287214d4dc9885ce4506457e12330baaf05e5d1bf29037ddcb3377b7704f2bd7 +size 75497472 diff --git a/params_shard_107.bin b/params_shard_107.bin new file mode 100644 index 0000000000000000000000000000000000000000..047355bb7b36f572e19c7f5a83d4f796acc5bdbc --- /dev/null +++ b/params_shard_107.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8c03e5dc9232108d478957e5e8b6a990e2d08f57fd1f746574a78d689ecaf123 +size 25165824 diff --git a/params_shard_108.bin b/params_shard_108.bin new file mode 100644 index 0000000000000000000000000000000000000000..715fd678c73426774c5929089a0a0f5046572e5d --- /dev/null +++ b/params_shard_108.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b5d6b8676c5fc0c607fa377ce55ed148454b3a615b43a794b7db10e35c56eee5 +size 150994944 diff --git a/params_shard_109.bin b/params_shard_109.bin new file mode 100644 index 0000000000000000000000000000000000000000..ce1df082507ab4ea18c8cab83a891fbe11a4f688 --- /dev/null +++ b/params_shard_109.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9ebdedcb0a510fb9361ba8ed57d3de9a5f59de79e6d49f19fa979dc461240b18 +size 301989888 diff --git a/params_shard_11.bin b/params_shard_11.bin new file mode 100644 index 0000000000000000000000000000000000000000..059b3fdf08cf884132cfb39c3ae7854b509e7bd1 --- /dev/null +++ b/params_shard_11.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:29d00ca49eacfb9e45b67eede2553a49a1b8dd373f862d4f3af0ea24ced09427 +size 25165824 diff --git a/params_shard_110.bin b/params_shard_110.bin new file mode 100644 index 0000000000000000000000000000000000000000..3c008fb53f56d3d004b46160d17217f6e7c72ec5 --- /dev/null +++ b/params_shard_110.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7228e0bfb02a2fc28910961f115740a8a2820e609c02612c766f71a20c801c3f +size 75497472 diff --git a/params_shard_111.bin b/params_shard_111.bin new file mode 100644 index 0000000000000000000000000000000000000000..7df8f67761727781e9faa972d0665a618c8e1ae2 --- /dev/null +++ b/params_shard_111.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:20a48f87b6e74bdc8a173ece8948c7a51eeeafeed7a67caed7092137da2138b3 +size 25165824 diff --git a/params_shard_112.bin b/params_shard_112.bin new file mode 100644 index 0000000000000000000000000000000000000000..e7474ba4ecad1606f4915293ee366906660c5400 --- /dev/null +++ b/params_shard_112.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:75a2ac9365ec97fb982adc314766c9140d5fd17e730025a0c9fc1a1bc2966e7a +size 25516032 diff --git a/params_shard_12.bin b/params_shard_12.bin new file mode 100644 index 0000000000000000000000000000000000000000..88a4b1a216ef38c70af651b5e047e9bb386735f8 --- /dev/null +++ b/params_shard_12.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:01c190c0c5a1bef0c37ac8729cd9ecf2decab6ebfcfbdd45c59cd6ed4d0f0b40 +size 150994944 diff --git a/params_shard_13.bin b/params_shard_13.bin new file mode 100644 index 0000000000000000000000000000000000000000..fb95e72baa99d172e5e7c7dca9ebb994224ca6a0 --- /dev/null +++ b/params_shard_13.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b0467c5f990c48a52cbb19956cdc0558b99826e6a81a97aa4c364e21f422b612 +size 301989888 diff --git a/params_shard_14.bin b/params_shard_14.bin new file mode 100644 index 0000000000000000000000000000000000000000..7940917c229afde7ecffde771170e5b08e299076 --- /dev/null +++ b/params_shard_14.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5a89b8c0d23829b44bd16af42cac3874c4746a0e2d19c4aaa5435c13e10227ff +size 75497472 diff --git a/params_shard_15.bin b/params_shard_15.bin new file mode 100644 index 0000000000000000000000000000000000000000..caa459f8440139dd0e76c4c52bfc65a26e4ee9ba --- /dev/null +++ b/params_shard_15.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0dcd93113065305440a919b3f36cc98266ddde041b28338459266fee7025e4f5 +size 25165824 diff --git a/params_shard_16.bin b/params_shard_16.bin new file mode 100644 index 0000000000000000000000000000000000000000..89f8d4f444a8656464db16cbb9f9d8842e742754 --- /dev/null +++ b/params_shard_16.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c21cac19e82636159bbde77f3442bfbb0a5964f5140f4ba8ef67615fe6e1defe +size 150994944 diff --git a/params_shard_17.bin b/params_shard_17.bin new file mode 100644 index 0000000000000000000000000000000000000000..cf11871b0eea828ab78c6fccf7888f280b82c1b5 --- /dev/null +++ b/params_shard_17.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3ef7a1d483a0b145914dad708e99343057ada0d578b44d46a9a01283026d786c +size 301989888 diff --git a/params_shard_18.bin b/params_shard_18.bin new file mode 100644 index 0000000000000000000000000000000000000000..70256f9513fec829f51c6be7e3e5878213093927 --- /dev/null +++ b/params_shard_18.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1556831a7c93a779fdc8f883a8e5a049853d92da60c3ac274c0b687f4e800189 +size 75497472 diff --git a/params_shard_19.bin b/params_shard_19.bin new file mode 100644 index 0000000000000000000000000000000000000000..ef691104c1e1d7b8bd99ee64817e768336e91092 --- /dev/null +++ b/params_shard_19.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:390402e35f7a78a1d6df66b48a6f0ebad17eb8d1f9dc63904a377915f2ac9e9c +size 25165824 diff --git a/params_shard_2.bin b/params_shard_2.bin new file mode 100644 index 0000000000000000000000000000000000000000..5857a28f1de6e308faaedb5ca58b595cce591d18 --- /dev/null +++ b/params_shard_2.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cc41cbb4742ec19569909c6e6508d110bcc69e0ec441c3d7187c11a4b5a8bc51 +size 301989888 diff --git a/params_shard_20.bin b/params_shard_20.bin new file mode 100644 index 0000000000000000000000000000000000000000..fe416b0b9784461719ced73358b01dc6a36814a3 --- /dev/null +++ b/params_shard_20.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:add8e17e8725e4af18982c37049fa00512bf6ea04dbfcbd985acaf15482b02ae +size 150994944 diff --git a/params_shard_21.bin b/params_shard_21.bin new file mode 100644 index 0000000000000000000000000000000000000000..dc2f5aa25765faf752f8084da496055cdf4757c8 --- /dev/null +++ b/params_shard_21.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ea9be0c048d0c25d8ed75c05ae54bc5489b637426079e48c0432d90b94af197b +size 301989888 diff --git a/params_shard_22.bin b/params_shard_22.bin new file mode 100644 index 0000000000000000000000000000000000000000..27d085d4c92587fcec4faa68f261a3e0cce17390 --- /dev/null +++ b/params_shard_22.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:98e0a596195c6d80747d6926194cb3999eef1564e85f1445b9c441a17b10f1cc +size 75497472 diff --git a/params_shard_23.bin b/params_shard_23.bin new file mode 100644 index 0000000000000000000000000000000000000000..f97d49d8cabb45f38d4c67119a1e843619e98da8 --- /dev/null +++ b/params_shard_23.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fd1c78dda5cc708f285290ebb651c02b49396d658b2c3b5c4f1a8f050c002333 +size 25165824 diff --git a/params_shard_24.bin b/params_shard_24.bin new file mode 100644 index 0000000000000000000000000000000000000000..54ada076fbfbd863eaea967f0c236cc75d335843 --- /dev/null +++ b/params_shard_24.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d410e3fdddfff45ab08d2d2553dd5a3f661b8fadb776bac7b5767b337ae1e188 +size 75497472 diff --git a/params_shard_25.bin b/params_shard_25.bin new file mode 100644 index 0000000000000000000000000000000000000000..35386904208eb9a150294fc98130c0e6e068cb06 --- /dev/null +++ b/params_shard_25.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d8ad13914ad3a7986a5a790678924fe6bb0969a8c08adb3ee2e486c69d20fbe1 +size 25165824 diff --git a/params_shard_26.bin b/params_shard_26.bin new file mode 100644 index 0000000000000000000000000000000000000000..db2baa54b47ba62071e13a2a196a4eec3a2363b7 --- /dev/null +++ b/params_shard_26.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7bfcdfa46774895c24bfbf6bdba3062d6542f14eab61151ad650042d5d789780 +size 150994944 diff --git a/params_shard_27.bin b/params_shard_27.bin new file mode 100644 index 0000000000000000000000000000000000000000..a69bdee3bfd65c63331f5dcb04bcfea62f8b4cad --- /dev/null +++ b/params_shard_27.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0fcc20b22b21ba8fe233a06da95d363b7eeebb3f0058c0a802b329ee2b994ad9 +size 301989888 diff --git a/params_shard_28.bin b/params_shard_28.bin new file mode 100644 index 0000000000000000000000000000000000000000..7df2620a22a2da27caa5cab07737ff1aa2f103ff --- /dev/null +++ b/params_shard_28.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bf74a5cc5663cc2d8b8319718ec726f733e01336ed8fe883579b2eae5b4834de +size 75497472 diff --git a/params_shard_29.bin b/params_shard_29.bin new file mode 100644 index 0000000000000000000000000000000000000000..1bcb0de01d4bb08bc29593df7b59cfad2782b751 --- /dev/null +++ b/params_shard_29.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:534a2e8eaea715211f7041303dbf5bf6e952518852df94636aab6c153d4e53f3 +size 25165824 diff --git a/params_shard_3.bin b/params_shard_3.bin new file mode 100644 index 0000000000000000000000000000000000000000..074d8b5656ae68639f1db0054bffae6bab6b689c --- /dev/null +++ b/params_shard_3.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c19d327b67db6cc77f0fb91db824e77ee1ae380c539c50bd15213c5cd89151a0 +size 75497472 diff --git a/params_shard_30.bin b/params_shard_30.bin new file mode 100644 index 0000000000000000000000000000000000000000..482f7ace87814b3a84f6724806b3850625d8880f --- /dev/null +++ b/params_shard_30.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b06db5be8d76176a2ab55ec01d09bd68fbbfb9f7f4cb64ca963957a0a107a823 +size 150994944 diff --git a/params_shard_31.bin b/params_shard_31.bin new file mode 100644 index 0000000000000000000000000000000000000000..847f0f81622699274c46ddb2094ae33f70ccd78d --- /dev/null +++ b/params_shard_31.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b3afdef4e2625d48e9354bed366c34ffda6e5aa0d36e588875925d82443836d6 +size 301989888 diff --git a/params_shard_32.bin b/params_shard_32.bin new file mode 100644 index 0000000000000000000000000000000000000000..4fc3f0d6faa2db289b8d87eb5ec537a93bc63d36 --- /dev/null +++ b/params_shard_32.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d37a270030226928d511c099d3d8814d7c6dc0f09ff5c48537a676601d7e5729 +size 75497472 diff --git a/params_shard_33.bin b/params_shard_33.bin new file mode 100644 index 0000000000000000000000000000000000000000..b053f6e73dff5366285a233abd018a0ff579a000 --- /dev/null +++ b/params_shard_33.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:da374a381380322dae5b3511d8b743fd8b4c68f5c3b91f4af83a0f7c8a9a8ae8 +size 25165824 diff --git a/params_shard_34.bin b/params_shard_34.bin new file mode 100644 index 0000000000000000000000000000000000000000..ef1cde62036dca5ce9290f57b2004ad82ee60793 --- /dev/null +++ b/params_shard_34.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:69b7ef4a75f5d9995bc4642159e01bebf9440ee9f30e16e8a34d641d0732cf78 +size 150994944 diff --git a/params_shard_35.bin b/params_shard_35.bin new file mode 100644 index 0000000000000000000000000000000000000000..8b01d48c916e41a385c6470ca02f39a21b1e39e8 --- /dev/null +++ b/params_shard_35.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ee20cb3527281637868920c8141965f999ff2389b680cf0b1c7d2a6d6b1a6d15 +size 301989888 diff --git a/params_shard_36.bin b/params_shard_36.bin new file mode 100644 index 0000000000000000000000000000000000000000..af90927ea9b0cb19e8201af9e788336b21f35d66 --- /dev/null +++ b/params_shard_36.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8b7c9499b3a3d7b36e3b455daae35763e31fe2b65aada9fcb78d13259658fab1 +size 75497472 diff --git a/params_shard_37.bin b/params_shard_37.bin new file mode 100644 index 0000000000000000000000000000000000000000..aed18ba70c065114d67a408036fb41c113cca4aa --- /dev/null +++ b/params_shard_37.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d69b893bd57be82d55da3d22e491a1de8b0e7a9e2aac91d4436d4803042b342e +size 25165824 diff --git a/params_shard_38.bin b/params_shard_38.bin new file mode 100644 index 0000000000000000000000000000000000000000..8cfc04fd26c1405fc1fa6f642a31f60e77cac8d3 --- /dev/null +++ b/params_shard_38.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cf6ac7817179a25f31aeeaeb5e67ee0b93b61d4eb2c3a546a0079f9854ad5759 +size 150994944 diff --git a/params_shard_39.bin b/params_shard_39.bin new file mode 100644 index 0000000000000000000000000000000000000000..af066a48b44fef6a549267998262a415ee1f9852 --- /dev/null +++ b/params_shard_39.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dfbd1a407d3ab5f73d2a0cfacb06a6c22f540161231b0726b7bdb19b4d60c5eb +size 301989888 diff --git a/params_shard_4.bin b/params_shard_4.bin new file mode 100644 index 0000000000000000000000000000000000000000..522c1e1932af4ca1d39907b3495e5f03fe5e1402 --- /dev/null +++ b/params_shard_4.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e07caf835643146f40c6ccf82fe2457166b0e90783e569be63034d769f5a9cb4 +size 150994944 diff --git a/params_shard_40.bin b/params_shard_40.bin new file mode 100644 index 0000000000000000000000000000000000000000..892d6826359ee08f86f2ddc95eccc0f57145aeef --- /dev/null +++ b/params_shard_40.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4da33c8f2213678587106b62e7ef1c81fbc73ef8b74801d4d6713f60a95dac9b +size 75497472 diff --git a/params_shard_41.bin b/params_shard_41.bin new file mode 100644 index 0000000000000000000000000000000000000000..fdb0f036c16ed22b8803e01a3d40844583c90161 --- /dev/null +++ b/params_shard_41.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8efbf896040fce1954589dcc96cea1fd5ce933fd80666929bba51fbac7821ced +size 25165824 diff --git a/params_shard_42.bin b/params_shard_42.bin new file mode 100644 index 0000000000000000000000000000000000000000..06a2c892d4b874bd72b8e111ccfc267b424e5d94 --- /dev/null +++ b/params_shard_42.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d1ed0331e4cac1476542466bd88d29a6d4a8aacbf5fb80c648e425061feda7c7 +size 150994944 diff --git a/params_shard_43.bin b/params_shard_43.bin new file mode 100644 index 0000000000000000000000000000000000000000..844a7ba04f00c8bbd89bcce17fb6ba4dd28716f4 --- /dev/null +++ b/params_shard_43.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:91038c599339a1ff4001a4e3b24fe70f938b02735465d6e56ff21d3205ba927d +size 301989888 diff --git a/params_shard_44.bin b/params_shard_44.bin new file mode 100644 index 0000000000000000000000000000000000000000..6f7c72e542e14b9564cbdf2858e4d80ef4e5291d --- /dev/null +++ b/params_shard_44.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9d2425f5b6c0ac4db94756c53f23684d8445bf84abb0a13b7eb4db679ba5f941 +size 75497472 diff --git a/params_shard_45.bin b/params_shard_45.bin new file mode 100644 index 0000000000000000000000000000000000000000..61e38e76f72395a3ba436093ccb54bea1101fcc5 --- /dev/null +++ b/params_shard_45.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:28ac530045d31baeba2eff3d1ae2f55cc38c630da57cc6461402766b99574c37 +size 25165824 diff --git a/params_shard_46.bin b/params_shard_46.bin new file mode 100644 index 0000000000000000000000000000000000000000..ec5e96451f64212a879ec1d5f125fec4f885d228 --- /dev/null +++ b/params_shard_46.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:286faef9fc46bfc543166e6e4c82c84b8b0a277f3a2f68287e37421947d8e382 +size 75497472 diff --git a/params_shard_47.bin b/params_shard_47.bin new file mode 100644 index 0000000000000000000000000000000000000000..1ff12f1a68767b3d8ef1f3c71444ee0ef577d443 --- /dev/null +++ b/params_shard_47.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3dd401456ba5c30016ae479cf1eceffd3d351c1d190d17fc5cb986c0e387d807 +size 25165824 diff --git a/params_shard_48.bin b/params_shard_48.bin new file mode 100644 index 0000000000000000000000000000000000000000..842a365132c32ae666d73584ea5117ca20dca81c --- /dev/null +++ b/params_shard_48.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:38811781c13bd11ae77e1dc1ad39c4d53fe684aca2e117af43c0197df40c1cd9 +size 150994944 diff --git a/params_shard_49.bin b/params_shard_49.bin new file mode 100644 index 0000000000000000000000000000000000000000..2e90e84e49c1510f38384ea40a50ec8096a528cf --- /dev/null +++ b/params_shard_49.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d053b4359e9bee631e7f67a5de07cbb84f13401d985ec8c9ef4e673a68d2dc86 +size 301989888 diff --git a/params_shard_5.bin b/params_shard_5.bin new file mode 100644 index 0000000000000000000000000000000000000000..9686cd3900b3ff0555fb580f0240241c9d79e72f --- /dev/null +++ b/params_shard_5.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a41b98dee8a13822067d013ced3f5c11af53c94a9e21a304a35e27629c061d5f +size 301989888 diff --git a/params_shard_50.bin b/params_shard_50.bin new file mode 100644 index 0000000000000000000000000000000000000000..a3d4d7c4aa5107ec87254f71823a3c0eed34d7f1 --- /dev/null +++ b/params_shard_50.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d211de1b398eade56183fabc94a0948ba1822debc0d0343af7a5a3b74d47ddcd +size 150994944 diff --git a/params_shard_51.bin b/params_shard_51.bin new file mode 100644 index 0000000000000000000000000000000000000000..6539e059fdb8f02d60879b204477411b01ecbc6e --- /dev/null +++ b/params_shard_51.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7084605bdc853d2f66ac7f97aefc61656e0fcc04f27180ecd3fc412b8d3db3ef +size 301989888 diff --git a/params_shard_52.bin b/params_shard_52.bin new file mode 100644 index 0000000000000000000000000000000000000000..f07b424b3b8605ace23f0e351bd5c122036ab9b4 --- /dev/null +++ b/params_shard_52.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d6adeb2540045251e9f93e207be8cd4f7aa90f04fc907f2d66a0b434a593f5e1 +size 75497472 diff --git a/params_shard_53.bin b/params_shard_53.bin new file mode 100644 index 0000000000000000000000000000000000000000..afffd0ef3e4d99c51c29820ad38841d3a90f53b6 --- /dev/null +++ b/params_shard_53.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2fee00bfb5928f7cc68a8b4b1df14896183dd3422bd744bea53d813b4e6177ad +size 25165824 diff --git a/params_shard_54.bin b/params_shard_54.bin new file mode 100644 index 0000000000000000000000000000000000000000..502b966644711af04dcc864b55b8e4d8a8cb6120 --- /dev/null +++ b/params_shard_54.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e62a594d19436fcaab4999a9520858d70456ce16a9d1c1f219143da5ba4effa9 +size 150994944 diff --git a/params_shard_55.bin b/params_shard_55.bin new file mode 100644 index 0000000000000000000000000000000000000000..5df0f20a045f66361754b5755ea9e774dd0249ab --- /dev/null +++ b/params_shard_55.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b831bb49921ddc15ddb9baa522e0ea111b90718daed80d049004dad4579b58b8 +size 301989888 diff --git a/params_shard_56.bin b/params_shard_56.bin new file mode 100644 index 0000000000000000000000000000000000000000..22daf91b947963cc8e275295ce36a92feabaef97 --- /dev/null +++ b/params_shard_56.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8cd4c95e5ad70fa25615ae102e1a287a5f990fc491f478e5e11a02dbb3c98812 +size 75497472 diff --git a/params_shard_57.bin b/params_shard_57.bin new file mode 100644 index 0000000000000000000000000000000000000000..29a14548358f1cfc900f2fc4693ce49b1be1ea9a --- /dev/null +++ b/params_shard_57.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0c6739149abb230f9b406ea2871e80027e62d34fe485f1aefafadf3017d7de38 +size 25165824 diff --git a/params_shard_58.bin b/params_shard_58.bin new file mode 100644 index 0000000000000000000000000000000000000000..2b978707410cfac6119818e04b59f752b429be32 --- /dev/null +++ b/params_shard_58.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:994d7d7689d612f75f3a7def08f484e8e89d2b1e6442d90430834daf45d184ce +size 150994944 diff --git a/params_shard_59.bin b/params_shard_59.bin new file mode 100644 index 0000000000000000000000000000000000000000..bfbae40a4ad0740245b6d51ef278580ddab755ca --- /dev/null +++ b/params_shard_59.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c6efbcc191c31c4130bdf6677c98464afc7a5ec3d6c6e8ac2c8c5f5203846212 +size 301989888 diff --git a/params_shard_6.bin b/params_shard_6.bin new file mode 100644 index 0000000000000000000000000000000000000000..0f2003637ce127d3a4ddbbb358ada8f57b40f8d3 --- /dev/null +++ b/params_shard_6.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:942909a50a46b0830c77db75ab81875dae43bba136b2f02a39444cec04cfb66d +size 75497472 diff --git a/params_shard_60.bin b/params_shard_60.bin new file mode 100644 index 0000000000000000000000000000000000000000..9c767fedfb699f49764b08635e83d43285d054b8 --- /dev/null +++ b/params_shard_60.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8bb1f3827c16a2819b762bf5ee23939fcfd94f75c0454fa386a1b0eec1cab1fb +size 75497472 diff --git a/params_shard_61.bin b/params_shard_61.bin new file mode 100644 index 0000000000000000000000000000000000000000..dbe98c79710bd023fb5bbc0d8b6b55221da135d9 --- /dev/null +++ b/params_shard_61.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bcac4cfbb5d530bcd4ae03073371566338a1cd06c485fce3d048491db2ff469d +size 25165824 diff --git a/params_shard_62.bin b/params_shard_62.bin new file mode 100644 index 0000000000000000000000000000000000000000..bfe08df4b29bed0d931f13dbf81eaf6a0a8a0575 --- /dev/null +++ b/params_shard_62.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:519da90b6dacf2ff510b09c8f7c90f8872911e7f5ace5bd3dbe0da860514bc51 +size 150994944 diff --git a/params_shard_63.bin b/params_shard_63.bin new file mode 100644 index 0000000000000000000000000000000000000000..d82f572e0cb1058142d014b8b3d7d28ec3737633 --- /dev/null +++ b/params_shard_63.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4dd38a7bce0422e3782b3fc76b3992ead5dcb01bd7eb91197b2eb995503f2aa5 +size 301989888 diff --git a/params_shard_64.bin b/params_shard_64.bin new file mode 100644 index 0000000000000000000000000000000000000000..f2588839326b56d182be8379636aea39e7847130 --- /dev/null +++ b/params_shard_64.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f769ef46e356a2df78823376e77c21e16b2964cdf978e50bf69898486186a100 +size 150994944 diff --git a/params_shard_65.bin b/params_shard_65.bin new file mode 100644 index 0000000000000000000000000000000000000000..4fe117f2b821040bc9e751213dc3c51fdfa79c3a --- /dev/null +++ b/params_shard_65.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fae9856e601a657f19e85d893e6dcda7fb084ce4482fbeae07fe150ce529d981 +size 301989888 diff --git a/params_shard_66.bin b/params_shard_66.bin new file mode 100644 index 0000000000000000000000000000000000000000..569ce1b9621158f46de610e231378121c786253f --- /dev/null +++ b/params_shard_66.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:db2aeb19abc8fc28ce4be08eba8894337db3ec22203af29f22d5ebcbd69a6ff0 +size 75497472 diff --git a/params_shard_67.bin b/params_shard_67.bin new file mode 100644 index 0000000000000000000000000000000000000000..2b99fe739aa9c057d0f3cfcef281b7efabc25ca2 --- /dev/null +++ b/params_shard_67.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ff47cd3441b30eea819f82186c9fdda67473b3939f2110bbbe9fc860d48d298b +size 25165824 diff --git a/params_shard_68.bin b/params_shard_68.bin new file mode 100644 index 0000000000000000000000000000000000000000..6f5bf0c75788c70e03cceb195ddd1b64700c9ed1 --- /dev/null +++ b/params_shard_68.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2d889fdeab1dee06fb3df91217989af520cb0604e2eaa1c3dec5919ae2985b5c +size 150994944 diff --git a/params_shard_69.bin b/params_shard_69.bin new file mode 100644 index 0000000000000000000000000000000000000000..083e78ca44b99ec8ee15179f8c56565b1d656a17 --- /dev/null +++ b/params_shard_69.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:18edd03727161f840b5df78e8843277fda0c2685ea72c07c633a12a828d3258f +size 301989888 diff --git a/params_shard_7.bin b/params_shard_7.bin new file mode 100644 index 0000000000000000000000000000000000000000..b562d79ea08ed7fba3cbfa71f7bae1e54598da61 --- /dev/null +++ b/params_shard_7.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:52bc234be328957c70f55a17899f73174bef188a20e3ed11f70e1642825e3bab +size 25165824 diff --git a/params_shard_70.bin b/params_shard_70.bin new file mode 100644 index 0000000000000000000000000000000000000000..011efcdde17cd01e4a83c40eb7694fd394e1174e --- /dev/null +++ b/params_shard_70.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:532fa7b4a81300cc4b66c446740a1550870a68e2149473012b0dccfb82c72b94 +size 75497472 diff --git a/params_shard_71.bin b/params_shard_71.bin new file mode 100644 index 0000000000000000000000000000000000000000..8ef0234023acaca3ded91ad461385f5565b43bde --- /dev/null +++ b/params_shard_71.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e127ff953f48f05b04286c4ca7fa3cdf31b3dd5fd0d340a3db4fd6866504d2ec +size 25165824 diff --git a/params_shard_72.bin b/params_shard_72.bin new file mode 100644 index 0000000000000000000000000000000000000000..caf6847ccabff0fbc17020d391ecf2401cdf535e --- /dev/null +++ b/params_shard_72.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:51a97442e7e71b98373579b2a7344e5073e51ca7037fe8cf8da11683f352fc26 +size 150994944 diff --git a/params_shard_73.bin b/params_shard_73.bin new file mode 100644 index 0000000000000000000000000000000000000000..f9cebf1823de4f5a5b839438f8b62cb91f24d757 --- /dev/null +++ b/params_shard_73.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bd5d405f932c6b9b6d1d513fbf697acaeac53c2159b3b079a56bb13e11a69a36 +size 301989888 diff --git a/params_shard_74.bin b/params_shard_74.bin new file mode 100644 index 0000000000000000000000000000000000000000..107fd68a4f969acb8491b7e72a7db343e8748266 --- /dev/null +++ b/params_shard_74.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7abe664d350ca5127c5e41d57e1b112cd527e0fdc3a62e68002cca5a83a34940 +size 75497472 diff --git a/params_shard_75.bin b/params_shard_75.bin new file mode 100644 index 0000000000000000000000000000000000000000..ce9c809634341e516c6ba40e5f565718fecea450 --- /dev/null +++ b/params_shard_75.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:79282d99e01c0518d0597373eb96408ea3d47e0a9c6a53f0cd61daa04ff2927a +size 25165824 diff --git a/params_shard_76.bin b/params_shard_76.bin new file mode 100644 index 0000000000000000000000000000000000000000..aee558e5f9f1287972f32f57ab3c0fd96a5f4216 --- /dev/null +++ b/params_shard_76.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e349bd38c6ffa22078e6218eafb1328beca870d9481f98a5e53dc1bace9d7160 +size 150994944 diff --git a/params_shard_77.bin b/params_shard_77.bin new file mode 100644 index 0000000000000000000000000000000000000000..5665cf1252e1e328c33fe8d3ab2f3a2d5f24cb22 --- /dev/null +++ b/params_shard_77.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4209c7272ff1ff4a59d7f93a106dfdf949968fff91952650033072607ccb9c13 +size 301989888 diff --git a/params_shard_78.bin b/params_shard_78.bin new file mode 100644 index 0000000000000000000000000000000000000000..62974807dbecc152b4ce3a03e22bac8587047ece --- /dev/null +++ b/params_shard_78.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8737439621dccfa7556d0067df7cc9762116ca396f2682ddca70c27f40a25227 +size 75497472 diff --git a/params_shard_79.bin b/params_shard_79.bin new file mode 100644 index 0000000000000000000000000000000000000000..793542d1775036f9581d58e930af1f8f76b59ceb --- /dev/null +++ b/params_shard_79.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:04e6d540724e767e313664854622fc5766c4168268b4e81435fe118e7c9146c8 +size 25165824 diff --git a/params_shard_8.bin b/params_shard_8.bin new file mode 100644 index 0000000000000000000000000000000000000000..5a4d16d5df247deb466295099f7942e2ef87fcda --- /dev/null +++ b/params_shard_8.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c78ce9637be3cd7ac67b06b1d935d7ca707b2d8938982e13d426ffa33651954f +size 150994944 diff --git a/params_shard_80.bin b/params_shard_80.bin new file mode 100644 index 0000000000000000000000000000000000000000..76cb856054f303fd5d1ad78fc2130773ce50ab9c --- /dev/null +++ b/params_shard_80.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:588e3d5eee9a14d981a66a4b23e0d8d8dfa58b557c51b9b7af28024b64352c49 +size 150994944 diff --git a/params_shard_81.bin b/params_shard_81.bin new file mode 100644 index 0000000000000000000000000000000000000000..650a15b9b346d375fc5a0f6c591a96c68a013631 --- /dev/null +++ b/params_shard_81.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b586cefa267016b653acd3933c8871328811350b3f477eb03c6288fdb64a2395 +size 301989888 diff --git a/params_shard_82.bin b/params_shard_82.bin new file mode 100644 index 0000000000000000000000000000000000000000..da6bc706fae75e1a96e6101a7cf71c43498611dc --- /dev/null +++ b/params_shard_82.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3b7a77604027f60542f499ff296799f931cd8ea66fd87e87f13f449ab0b7697c +size 75497472 diff --git a/params_shard_83.bin b/params_shard_83.bin new file mode 100644 index 0000000000000000000000000000000000000000..d53a5966ee27d2f3880f02fa38c8da4aecc0d54b --- /dev/null +++ b/params_shard_83.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2300d7d1675f1089ae52a1f829fc10c9697b80b6124b27cd60df66394e9640f5 +size 25165824 diff --git a/params_shard_84.bin b/params_shard_84.bin new file mode 100644 index 0000000000000000000000000000000000000000..471ffc79b32c399ac872ece617bcffaca6d25a3e --- /dev/null +++ b/params_shard_84.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d2e1162410b906ffca6551b2231deb870f00eeb77f638a0b55fbcd00958dd23b +size 150994944 diff --git a/params_shard_85.bin b/params_shard_85.bin new file mode 100644 index 0000000000000000000000000000000000000000..78b4bca8c9e51c90130eeb3415d592ec2fc290a9 --- /dev/null +++ b/params_shard_85.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8188bd832ec02436275a78cf6d822dc2ea067d2888199db929ff2f5e0e1dc65d +size 301989888 diff --git a/params_shard_86.bin b/params_shard_86.bin new file mode 100644 index 0000000000000000000000000000000000000000..bed32fc46c1f98b305d60b606b5c57b3114cb835 --- /dev/null +++ b/params_shard_86.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f00e3483e1509686013827eda7dcaed6ebcd64e196611d95519c16812dcec2ae +size 75497472 diff --git a/params_shard_87.bin b/params_shard_87.bin new file mode 100644 index 0000000000000000000000000000000000000000..021ba60aaea0ad7ce5fddac4c6cfe6f1a7aa154b --- /dev/null +++ b/params_shard_87.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1069e1f7591fd5dd6505b54473f6286c9b30e07d53cd5ba44e0efc3e5bfcf494 +size 25165824 diff --git a/params_shard_88.bin b/params_shard_88.bin new file mode 100644 index 0000000000000000000000000000000000000000..6210f3cac33695561a4e2bcb8ac2ff3df233d25c --- /dev/null +++ b/params_shard_88.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e15b1f0ed835d49fba151fc5a438b6440d94ed66095e44dc93feed3321e7a55c +size 150994944 diff --git a/params_shard_89.bin b/params_shard_89.bin new file mode 100644 index 0000000000000000000000000000000000000000..1da510cdd089567c9937f75cb3add5b144f55c80 --- /dev/null +++ b/params_shard_89.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:491d6a2f325d728d9344db64826650e80f4c39c8b1216d4d430d11bf1aa9c09a +size 301989888 diff --git a/params_shard_9.bin b/params_shard_9.bin new file mode 100644 index 0000000000000000000000000000000000000000..16482108999523da4a589b2ee5f4d62f90a15fa2 --- /dev/null +++ b/params_shard_9.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:30cfac1ff3056b73d3b1b1cf6d1ba9854a069e5d90af24ff102ab1b43911ba40 +size 301989888 diff --git a/params_shard_90.bin b/params_shard_90.bin new file mode 100644 index 0000000000000000000000000000000000000000..ee8820cb9284a0352f86acf5ea0b96797025fb0d --- /dev/null +++ b/params_shard_90.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5b1422a035616d1ddc2a9e964cd23dca4a6c21b7a1307a27c363f287bdf07a20 +size 75497472 diff --git a/params_shard_91.bin b/params_shard_91.bin new file mode 100644 index 0000000000000000000000000000000000000000..cc536d6be2c875772073b8fb8674dec5ee5f3c5f --- /dev/null +++ b/params_shard_91.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:50281adbcefe4e70248b7c5e1d532803ab423860f0279b2feeed58b47939a833 +size 25165824 diff --git a/params_shard_92.bin b/params_shard_92.bin new file mode 100644 index 0000000000000000000000000000000000000000..5b1e20433332e0b271e5dcba9006cf8e9f174fab --- /dev/null +++ b/params_shard_92.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6a2f9a54346321632ceb21276014c2c20ae0879c62e7d76d61cb1009d29e1e9e +size 150994944 diff --git a/params_shard_93.bin b/params_shard_93.bin new file mode 100644 index 0000000000000000000000000000000000000000..a46a04a1a175c2c213c61b858d3a898460aa819f --- /dev/null +++ b/params_shard_93.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:45452afe8f512f5085b214d50cc009261324a5ff94aae291da3360f8a560d8a3 +size 301989888 diff --git a/params_shard_94.bin b/params_shard_94.bin new file mode 100644 index 0000000000000000000000000000000000000000..a0257cbe8cac7daa267d0643831627fea5224605 --- /dev/null +++ b/params_shard_94.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:894d1562819fcf9642e5f791ca83132ac2bd296ceb28642c75727a599ec8f1a6 +size 75497472 diff --git a/params_shard_95.bin b/params_shard_95.bin new file mode 100644 index 0000000000000000000000000000000000000000..2f131d1854c32b772feb37ca8551285742ac5a79 --- /dev/null +++ b/params_shard_95.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6ed958b731fd10a3313259d9eb68f9774e505afd818d5310dad60887fb6fd22a +size 25165824 diff --git a/params_shard_96.bin b/params_shard_96.bin new file mode 100644 index 0000000000000000000000000000000000000000..f522c8659d936d35905ccd98541d4bafc20ded23 --- /dev/null +++ b/params_shard_96.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:afddf3ea23c2e303625867474e73e419482116159490d79654972aebb1960cab +size 75497472 diff --git a/params_shard_97.bin b/params_shard_97.bin new file mode 100644 index 0000000000000000000000000000000000000000..eb4e8c79ce88ca67341e99fac6d5079052e3e75e --- /dev/null +++ b/params_shard_97.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6ac10b35bfe169167fdff2adb4b2f33155b4025a5ced83ef5831533b07a4bb52 +size 25165824 diff --git a/params_shard_98.bin b/params_shard_98.bin new file mode 100644 index 0000000000000000000000000000000000000000..6ca3d45ff1b2c2469967830e39355731ce209c72 --- /dev/null +++ b/params_shard_98.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:18ce002b1349a0ca29ab043b0b92a5626cf9d924a59a0a47df3e1dba8af6360e +size 150994944 diff --git a/params_shard_99.bin b/params_shard_99.bin new file mode 100644 index 0000000000000000000000000000000000000000..5e1a946134f758a976d55f9864489e641173ca29 --- /dev/null +++ b/params_shard_99.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:de6359104a2c59fae19876d3ac7e593a52dfac606569ef51ae08420d5c120868 +size 301989888 diff --git a/tokenizer.json b/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..e7caf9f7f6e58c4bd4563dc5b2fd2e596f0ad070 --- /dev/null +++ b/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c15eb04bc5ad609fb26533e8525302c5640a945e5f67f65b7c849900acda7d99 +size 17518497 diff --git a/tokenizer.model b/tokenizer.model new file mode 100644 index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8 --- /dev/null +++ b/tokenizer.model @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2 +size 4241003 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..035fa0e287781a32a1bb2ff9f5cfd6cb5522be70 --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,1517 @@ +{ + "add_bos_token": true, + "add_eos_token": false, + "added_tokens_decoder": { + "0": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "1": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "2": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "3": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "4": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "5": { + "content": "<2mass>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "6": { + "content": "[@BOS@]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "7": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "8": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "9": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "10": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "11": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "12": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "13": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "14": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "15": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "16": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "17": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "18": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "19": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "20": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "21": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "22": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "23": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "24": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "25": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "26": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "27": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "28": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "29": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "30": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "31": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "32": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "33": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "34": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "35": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "36": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "37": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "38": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "39": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "40": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "41": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "42": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "43": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "44": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "45": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "46": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "47": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "48": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "49": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "50": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "51": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "52": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "53": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "54": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "55": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "56": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "57": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "58": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "59": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "60": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "61": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "62": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "63": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "64": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "65": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "66": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "67": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "68": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "69": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "70": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "71": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "72": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "73": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "74": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "75": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "76": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "77": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "78": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "79": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "80": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "81": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "82": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "83": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "84": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "85": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "86": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "87": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "88": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "89": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "90": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "91": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "92": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "93": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "94": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "95": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "96": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "97": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "98": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "99": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "100": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "101": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "102": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "103": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "104": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "105": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "106": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "107": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "108": { + "content": "\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "109": { + "content": "\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "110": { + "content": "\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "111": { + "content": "\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "112": { + "content": "\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "113": { + "content": "\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "114": { + "content": "\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "115": { + "content": "\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "116": { + "content": "\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "117": { + "content": "\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "118": { + "content": "\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "119": { + "content": "\n\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "120": { + "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "121": { + "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "122": { + "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "123": { + "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "124": { + "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "125": { + "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "126": { + "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "127": { + "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "128": { + "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "129": { + "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "130": { + "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "131": { + "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "132": { + "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "133": { + "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "134": { + "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "135": { + "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "136": { + "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "137": { + "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "138": { + "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "169": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "170": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "172": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "173": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "174": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "175": { + "content": "
", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "171": { + "content": "
", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "176": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "177": { + "content": "
", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "178": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "179": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "180": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "181": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "182": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "183": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "184": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "185": { + "content": "

", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "186": { + "content": "

", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "187": { + "content": "

", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "188": { + "content": "

", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "189": { + "content": "

", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "190": { + "content": "
", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "191": { + "content": "
", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "192": { + "content": "
", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "193": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "194": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "195": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "196": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "197": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "198": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "199": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "200": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "201": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "202": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "203": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "204": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "205": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "206": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "207": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "208": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "209": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "210": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "211": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "212": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "213": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "214": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "215": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "216": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + } + }, + "additional_special_tokens": [ + "", + "" + ], + "bos_token": "", + "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '' + role + '\n' + message['content'] | trim + '\n' }}{% endfor %}{% if add_generation_prompt %}{{'model\n'}}{% endif %}", + "clean_up_tokenization_spaces": false, + "eos_token": "", + "model_max_length": 1000000000000000019884624838656, + "pad_token": "", + "sp_model_kwargs": {}, + "spaces_between_special_tokens": false, + "tokenizer_class": "GemmaTokenizer", + "unk_token": "", + "use_default_system_prompt": false +}