Upload MPTForCausalLM

config.json

@@ -0,0 +1,53 @@
+{
+  "_name_or_path": "mosaicml/mpt-7b-8k-instruct",
+  "architectures": [
+    "MPTForCausalLM"
+  ],
+  "attn_config": {
+    "alibi": true,
+    "alibi_bias_max": 8,
+    "attn_impl": "torch",
+    "attn_pdrop": 0,
+    "attn_type": "multihead_attention",
+    "attn_uses_sequence_id": false,
+    "clip_qkv": null,
+    "prefix_lm": false,
+    "qk_ln": false,
+    "softmax_scale": null
+  },
+  "auto_map": {
+    "AutoConfig": "mosaicml/mpt-7b-8k-instruct--configuration_mpt.MPTConfig",
+    "AutoModelForCausalLM": "mosaicml/mpt-7b-8k-instruct--modeling_mpt.MPTForCausalLM"
+  },
+  "d_model": 4096,
+  "emb_pdrop": 0,
+  "embedding_fraction": 1.0,
+  "expansion_ratio": 4,
+  "init_config": {
+    "emb_init_std": null,
+    "emb_init_uniform_lim": null,
+    "fan_mode": "fan_in",
+    "init_div_is_residual": true,
+    "init_gain": 0.0,
+    "init_nonlinearity": "relu",
+    "init_std": null,
+    "name": "kaiming_normal_",
+    "verbose": 0
+  },
+  "init_device": "cpu",
+  "learned_pos_emb": true,
+  "logit_scale": null,
+  "max_seq_len": 8192,
+  "model_type": "mpt",
+  "n_heads": 32,
+  "n_layers": 32,
+  "no_bias": true,
+  "norm_type": "low_precision_layernorm",
+  "resid_pdrop": 0,
+  "tokenizer_name": "EleutherAI/gpt-neox-20b",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.32.0.dev0",
+  "use_cache": false,
+  "verbose": 0,
+  "vocab_size": 50432
+}

generation_config.json

@@ -0,0 +1,5 @@
+{
+  "_from_model_config": true,
+  "transformers_version": "4.32.0.dev0",
+  "use_cache": false
+}

model-00001-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa4f226ab3582a701902577c6df6e173d9fff244af7fa81ebd55ab776cccec51
+size 1889602248

model-00002-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:20f9502765398bb6f2a86d690c8dd7af42e43afa34c78a0273f8f1b733749d81
+size 1879133464

model-00003-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a6d144b2faa13f492869e25a1a6571b350695a4f77f5acfb4d5ad7dab0fcd4b
+size 1979788480

model-00004-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf866394f5777df04056dbbb2cd2615f728c7e767cdd2c6fe3d2d03b8cf3cf82
+size 1912687928

model-00005-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9a9d3a0567f7d504fc9f7aeb8ddf35ceb8c1059622a6ef89707de8fafc3acde
+size 1879125192

model-00006-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69e81367b596dd68958463490fe48c0e8fb23b81365fc0401f66be4db5907188
+size 1879133496

model-00007-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98ea48e16b5513f0490a8fab66e0afa0104792408483d9a28cc8f10f7a906afb
+size 1879125072

model.safetensors.index.json

@@ -0,0 +1,201 @@
+{
+  "metadata": {
+    "total_size": 13298573312
+  },
+  "weight_map": {
+    "transformer.blocks.0.attn.Wqkv.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.0.attn.out_proj.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.0.ffn.down_proj.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.0.ffn.up_proj.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.0.norm_1.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.0.norm_2.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.1.attn.Wqkv.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.1.attn.out_proj.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.1.ffn.down_proj.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.1.ffn.up_proj.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.1.norm_1.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.1.norm_2.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.10.attn.Wqkv.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.10.attn.out_proj.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.10.ffn.down_proj.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.10.ffn.up_proj.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.10.norm_1.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.10.norm_2.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.11.attn.Wqkv.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.11.attn.out_proj.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.11.ffn.down_proj.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.11.ffn.up_proj.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.11.norm_1.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.11.norm_2.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.12.attn.Wqkv.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.12.attn.out_proj.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.12.ffn.down_proj.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.12.ffn.up_proj.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.12.norm_1.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.12.norm_2.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.13.attn.Wqkv.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.13.attn.out_proj.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.13.ffn.down_proj.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.13.ffn.up_proj.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.13.norm_1.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.13.norm_2.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.14.attn.Wqkv.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.14.attn.out_proj.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.14.ffn.down_proj.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.14.ffn.up_proj.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.14.norm_1.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.14.norm_2.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.15.attn.Wqkv.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.15.attn.out_proj.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.15.ffn.down_proj.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.15.ffn.up_proj.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.15.norm_1.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.15.norm_2.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.16.attn.Wqkv.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.16.attn.out_proj.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.16.ffn.down_proj.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.16.ffn.up_proj.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.16.norm_1.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.16.norm_2.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.17.attn.Wqkv.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.17.attn.out_proj.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.17.ffn.down_proj.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.17.ffn.up_proj.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.17.norm_1.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.17.norm_2.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.18.attn.Wqkv.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.18.attn.out_proj.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.18.ffn.down_proj.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.18.ffn.up_proj.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.18.norm_1.weight": "model-00004-of-00007.safetensors",
+    "transformer.blocks.18.norm_2.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.19.attn.Wqkv.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.19.attn.out_proj.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.19.ffn.down_proj.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.19.ffn.up_proj.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.19.norm_1.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.19.norm_2.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.2.attn.Wqkv.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.2.attn.out_proj.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.2.ffn.down_proj.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.2.ffn.up_proj.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.2.norm_1.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.2.norm_2.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.20.attn.Wqkv.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.20.attn.out_proj.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.20.ffn.down_proj.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.20.ffn.up_proj.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.20.norm_1.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.20.norm_2.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.21.attn.Wqkv.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.21.attn.out_proj.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.21.ffn.down_proj.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.21.ffn.up_proj.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.21.norm_1.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.21.norm_2.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.22.attn.Wqkv.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.22.attn.out_proj.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.22.ffn.down_proj.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.22.ffn.up_proj.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.22.norm_1.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.22.norm_2.weight": "model-00005-of-00007.safetensors",
+    "transformer.blocks.23.attn.Wqkv.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.23.attn.out_proj.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.23.ffn.down_proj.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.23.ffn.up_proj.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.23.norm_1.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.23.norm_2.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.24.attn.Wqkv.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.24.attn.out_proj.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.24.ffn.down_proj.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.24.ffn.up_proj.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.24.norm_1.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.24.norm_2.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.25.attn.Wqkv.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.25.attn.out_proj.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.25.ffn.down_proj.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.25.ffn.up_proj.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.25.norm_1.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.25.norm_2.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.26.attn.Wqkv.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.26.attn.out_proj.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.26.ffn.down_proj.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.26.ffn.up_proj.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.26.norm_1.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.26.norm_2.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.27.attn.Wqkv.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.27.attn.out_proj.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.27.ffn.down_proj.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.27.ffn.up_proj.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.27.norm_1.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.27.norm_2.weight": "model-00006-of-00007.safetensors",
+    "transformer.blocks.28.attn.Wqkv.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.28.attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.28.ffn.down_proj.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.28.ffn.up_proj.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.28.norm_1.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.28.norm_2.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.29.attn.Wqkv.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.29.attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.29.ffn.down_proj.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.29.ffn.up_proj.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.29.norm_1.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.29.norm_2.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.3.attn.Wqkv.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.3.attn.out_proj.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.3.ffn.down_proj.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.3.ffn.up_proj.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.3.norm_1.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.3.norm_2.weight": "model-00001-of-00007.safetensors",
+    "transformer.blocks.30.attn.Wqkv.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.30.attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.30.ffn.down_proj.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.30.ffn.up_proj.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.30.norm_1.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.30.norm_2.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.31.attn.Wqkv.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.31.attn.out_proj.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.31.ffn.down_proj.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.31.ffn.up_proj.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.31.norm_1.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.31.norm_2.weight": "model-00007-of-00007.safetensors",
+    "transformer.blocks.4.attn.Wqkv.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.4.attn.out_proj.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.4.ffn.down_proj.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.4.ffn.up_proj.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.4.norm_1.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.4.norm_2.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.5.attn.Wqkv.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.5.attn.out_proj.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.5.ffn.down_proj.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.5.ffn.up_proj.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.5.norm_1.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.5.norm_2.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.6.attn.Wqkv.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.6.attn.out_proj.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.6.ffn.down_proj.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.6.ffn.up_proj.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.6.norm_1.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.6.norm_2.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.7.attn.Wqkv.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.7.attn.out_proj.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.7.ffn.down_proj.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.7.ffn.up_proj.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.7.norm_1.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.7.norm_2.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.8.attn.Wqkv.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.8.attn.out_proj.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.8.ffn.down_proj.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.8.ffn.up_proj.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.8.norm_1.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.8.norm_2.weight": "model-00002-of-00007.safetensors",
+    "transformer.blocks.9.attn.Wqkv.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.9.attn.out_proj.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.9.ffn.down_proj.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.9.ffn.up_proj.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.9.norm_1.weight": "model-00003-of-00007.safetensors",
+    "transformer.blocks.9.norm_2.weight": "model-00003-of-00007.safetensors",
+    "transformer.norm_f.weight": "model-00007-of-00007.safetensors",
+    "transformer.wte.weight": "model-00001-of-00007.safetensors"
+  }
+}