Adun commited on
Commit
4a76e31
1 Parent(s): 5105a17

Initial Version

Browse files
Files changed (43) hide show
  1. config.json +29 -0
  2. generation_config.json +6 -0
  3. model-00001-of-00040.safetensors +3 -0
  4. model-00002-of-00040.safetensors +3 -0
  5. model-00003-of-00040.safetensors +3 -0
  6. model-00004-of-00040.safetensors +3 -0
  7. model-00005-of-00040.safetensors +3 -0
  8. model-00006-of-00040.safetensors +3 -0
  9. model-00007-of-00040.safetensors +3 -0
  10. model-00008-of-00040.safetensors +3 -0
  11. model-00009-of-00040.safetensors +3 -0
  12. model-00010-of-00040.safetensors +3 -0
  13. model-00011-of-00040.safetensors +3 -0
  14. model-00012-of-00040.safetensors +3 -0
  15. model-00013-of-00040.safetensors +3 -0
  16. model-00014-of-00040.safetensors +3 -0
  17. model-00015-of-00040.safetensors +3 -0
  18. model-00016-of-00040.safetensors +3 -0
  19. model-00017-of-00040.safetensors +3 -0
  20. model-00018-of-00040.safetensors +3 -0
  21. model-00019-of-00040.safetensors +3 -0
  22. model-00020-of-00040.safetensors +3 -0
  23. model-00021-of-00040.safetensors +3 -0
  24. model-00022-of-00040.safetensors +3 -0
  25. model-00023-of-00040.safetensors +3 -0
  26. model-00024-of-00040.safetensors +3 -0
  27. model-00025-of-00040.safetensors +3 -0
  28. model-00026-of-00040.safetensors +3 -0
  29. model-00027-of-00040.safetensors +3 -0
  30. model-00028-of-00040.safetensors +3 -0
  31. model-00029-of-00040.safetensors +3 -0
  32. model-00030-of-00040.safetensors +3 -0
  33. model-00031-of-00040.safetensors +3 -0
  34. model-00032-of-00040.safetensors +3 -0
  35. model-00033-of-00040.safetensors +3 -0
  36. model-00034-of-00040.safetensors +3 -0
  37. model-00035-of-00040.safetensors +3 -0
  38. model-00036-of-00040.safetensors +3 -0
  39. model-00037-of-00040.safetensors +3 -0
  40. model-00038-of-00040.safetensors +3 -0
  41. model-00039-of-00040.safetensors +3 -0
  42. model-00040-of-00040.safetensors +3 -0
  43. model.safetensors.index.json +298 -0
config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/content/drive/MyDrive/BaseModel/openthaigpt-1.0.0-7b-chat",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 4096,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 11008,
14
+ "max_position_embeddings": 4096,
15
+ "mlp_bias": false,
16
+ "model_type": "llama",
17
+ "num_attention_heads": 32,
18
+ "num_hidden_layers": 32,
19
+ "num_key_value_heads": 32,
20
+ "pretraining_tp": 1,
21
+ "rms_norm_eps": 1e-05,
22
+ "rope_scaling": null,
23
+ "rope_theta": 10000.0,
24
+ "tie_word_embeddings": false,
25
+ "torch_dtype": "float16",
26
+ "transformers_version": "4.41.2",
27
+ "use_cache": false,
28
+ "vocab_size": 41088
29
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "transformers_version": "4.41.2"
6
+ }
model-00001-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:40ded83bcd9a174a40dc24fec109894c3fc1fe8756004ccce6d98c9f39ff0c33
3
+ size 370147576
model-00002-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:16def2c80e8abe4e90629cb4762153c2fa129099f9e33f1acffa83203e10fcc3
3
+ size 371213216
model-00003-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:633d3f2b3be5c022221cf7add2d85daf59eca23cfb3ee5582f65dab5e1deba46
3
+ size 314573512
model-00004-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4ab137aff77743b3340531c69a71d227f15a5262d73ba9c76a05dba8135b9816
3
+ size 314590112
model-00005-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:65c19b5d14e2fc81445e8522d40d0c21d75b456cd7e81a06bf57ad8916fafd68
3
+ size 314590112
model-00006-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a27aa4ff45d121c054dd6082c7dbdcabf3b156aed857d5eef14cd94ab362be1f
3
+ size 371213216
model-00007-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:35af34d8fdce6b5ca8b0ade685732862c900435c9bdb40de9778ea8066e62143
3
+ size 371213216
model-00008-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d4c5132daf135a3d4d16cb01b8b08af59a4b79a660430ae4082ae584c20c866
3
+ size 371213216
model-00009-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e3e90e600edceda8aebde35da664b0706f4c3c8e6b86135070803b76383039e9
3
+ size 371213216
model-00010-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8311ec67af9e0dc709bcb217b846a2e2897b1bf09b934233a187adbd1eac3929
3
+ size 314573512
model-00011-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee40eecbc654d34016205c35bf72787add52f55d90f36234ffbf3f5a4e82718c
3
+ size 314590112
model-00012-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e9379598a309eba775568c93b56fde446a4550f21caf5d60c5965a612d939947
3
+ size 314590112
model-00013-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:088b1a3508f4ef31ed27a5c096d2409cb0a78441f25cb86c21c82503150cc7d6
3
+ size 371213224
model-00014-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4c30a49d803aeebd353173c82623a9c47842005c44d2facef5c3678e1955463
3
+ size 371213224
model-00015-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94725f1310e6879d242c423cbbc0823ac4148e35dae2ebd070ec927ffc6ed63f
3
+ size 371213224
model-00016-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1abc7c03054884d0fe427217d7357bc09ba0528f9c57e9a18e6e9cf7a5c6f0af
3
+ size 371213224
model-00017-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:41b3c2e5da6410d810d0a9da6aa334d66b1fb49d35029d25dd7330e455b433eb
3
+ size 314573520
model-00018-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5189e45865a4bf297bcb082c3c36a88304fe1d6fe03a087fb3ec6a7e61b04d75
3
+ size 314590120
model-00019-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bbba25d69ef8d533d2c1f2194c63ec6fb4ff4f187e7d53ef201112902d58bf5f
3
+ size 314590120
model-00020-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3ce1132fb4ebec07e43f53add5f05e70705fbc5ec37a74705b2e6a211240baab
3
+ size 371213224
model-00021-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:550f4fbc7e14bd77634a99c1ea3b7efac8cbfb8fb2856432753f8020c8f526c3
3
+ size 371213224
model-00022-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a759f58acc9565de915772d900d3d0af0294fbb32a87440f41b69e8ed6f1d0b
3
+ size 371213224
model-00023-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2bea8b7431c8e13fe4c0899b6f7f733117c2ed61702a8a4c146fc97a97bfb22b
3
+ size 371213224
model-00024-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f85338e7090a323464581ffa982197160501b1e5eb5eafb7831a08c8421ec7e1
3
+ size 314573520
model-00025-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:79b2cedf6df8499cb3b7450adc04ff559a0800d78a79aa77ba4873b3286fa71c
3
+ size 314590120
model-00026-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:32285d1b6604bac867e682552dd789744c8d0bd34c1921fb2277d77aed3c58f4
3
+ size 314590120
model-00027-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81aa7a0d8d075f76e16869daae867c62c02255c1baf8be1884349b339126ef1c
3
+ size 371213224
model-00028-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:be7a90d310382662f67b426c12dcd1162df7146f183bcd40a97c32dd63d3dab6
3
+ size 371213224
model-00029-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7980c0aa2fb18cd031d79445ea19e183bad313822a312365039fc891be3dceac
3
+ size 371213224
model-00030-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e2572fb06264521a003b65d978e837290d8d078fe283f49b0c28b286278b174
3
+ size 371213224
model-00031-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c791b9e8789abe997349f850e61e3b4c7e5989ffdcff3dec8ada612d59089004
3
+ size 314573520
model-00032-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:952112364adc218186a4625989cdc5a1ab387097586035f1ff2d153c16763ba1
3
+ size 314590120
model-00033-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07e0fac7dc950b6b419b0c5fb8a8737d914cd61e1015b0cffd2cd04a6b20844f
3
+ size 314590120
model-00034-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b22a6a6d2b53c8c75ed34c7a31b007baa9db2c0f0b46a3d5a26b95dacccdfd8c
3
+ size 371213224
model-00035-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:19db0236224ae58cf597244765ac715608af8522366f671471f1d75fe87fd475
3
+ size 371213224
model-00036-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6ca3d5a719e2490c295855b4b1a89696d3c940eae05467895d5dc11db855dbd
3
+ size 371213224
model-00037-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:690e4710399c539420821d460807b3db1f01905a3496e9956435b7abee4fa0a2
3
+ size 371213224
model-00038-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca3138a88c16169db471936b9488e764c75d5a07df724ac5a0fa8d1e77f0c5ec
3
+ size 314573520
model-00039-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9d6093a108a4ff0bb2de8d389044163c2c848a8256e2f805337b346626950503
3
+ size 90202560
model-00040-of-00040.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f48d2cd80cb7cb6fb89df34f84fd560f693312dcb3e48d747e2d61e39b4e0b21
3
+ size 336593024
model.safetensors.index.json ADDED
@@ -0,0 +1,298 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 13625729024
4
+ },
5
+ "weight_map": {
6
+ "lm_head.weight": "model-00040-of-00040.safetensors",
7
+ "model.embed_tokens.weight": "model-00001-of-00040.safetensors",
8
+ "model.layers.0.input_layernorm.weight": "model-00002-of-00040.safetensors",
9
+ "model.layers.0.mlp.down_proj.weight": "model-00002-of-00040.safetensors",
10
+ "model.layers.0.mlp.gate_proj.weight": "model-00002-of-00040.safetensors",
11
+ "model.layers.0.mlp.up_proj.weight": "model-00002-of-00040.safetensors",
12
+ "model.layers.0.post_attention_layernorm.weight": "model-00002-of-00040.safetensors",
13
+ "model.layers.0.self_attn.k_proj.weight": "model-00002-of-00040.safetensors",
14
+ "model.layers.0.self_attn.o_proj.weight": "model-00002-of-00040.safetensors",
15
+ "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00040.safetensors",
16
+ "model.layers.0.self_attn.v_proj.weight": "model-00002-of-00040.safetensors",
17
+ "model.layers.1.input_layernorm.weight": "model-00004-of-00040.safetensors",
18
+ "model.layers.1.mlp.down_proj.weight": "model-00004-of-00040.safetensors",
19
+ "model.layers.1.mlp.gate_proj.weight": "model-00003-of-00040.safetensors",
20
+ "model.layers.1.mlp.up_proj.weight": "model-00003-of-00040.safetensors",
21
+ "model.layers.1.post_attention_layernorm.weight": "model-00004-of-00040.safetensors",
22
+ "model.layers.1.self_attn.k_proj.weight": "model-00003-of-00040.safetensors",
23
+ "model.layers.1.self_attn.o_proj.weight": "model-00003-of-00040.safetensors",
24
+ "model.layers.1.self_attn.q_proj.weight": "model-00003-of-00040.safetensors",
25
+ "model.layers.1.self_attn.v_proj.weight": "model-00003-of-00040.safetensors",
26
+ "model.layers.10.input_layernorm.weight": "model-00014-of-00040.safetensors",
27
+ "model.layers.10.mlp.down_proj.weight": "model-00014-of-00040.safetensors",
28
+ "model.layers.10.mlp.gate_proj.weight": "model-00014-of-00040.safetensors",
29
+ "model.layers.10.mlp.up_proj.weight": "model-00014-of-00040.safetensors",
30
+ "model.layers.10.post_attention_layernorm.weight": "model-00014-of-00040.safetensors",
31
+ "model.layers.10.self_attn.k_proj.weight": "model-00013-of-00040.safetensors",
32
+ "model.layers.10.self_attn.o_proj.weight": "model-00014-of-00040.safetensors",
33
+ "model.layers.10.self_attn.q_proj.weight": "model-00013-of-00040.safetensors",
34
+ "model.layers.10.self_attn.v_proj.weight": "model-00013-of-00040.safetensors",
35
+ "model.layers.11.input_layernorm.weight": "model-00015-of-00040.safetensors",
36
+ "model.layers.11.mlp.down_proj.weight": "model-00015-of-00040.safetensors",
37
+ "model.layers.11.mlp.gate_proj.weight": "model-00015-of-00040.safetensors",
38
+ "model.layers.11.mlp.up_proj.weight": "model-00015-of-00040.safetensors",
39
+ "model.layers.11.post_attention_layernorm.weight": "model-00015-of-00040.safetensors",
40
+ "model.layers.11.self_attn.k_proj.weight": "model-00014-of-00040.safetensors",
41
+ "model.layers.11.self_attn.o_proj.weight": "model-00015-of-00040.safetensors",
42
+ "model.layers.11.self_attn.q_proj.weight": "model-00014-of-00040.safetensors",
43
+ "model.layers.11.self_attn.v_proj.weight": "model-00015-of-00040.safetensors",
44
+ "model.layers.12.input_layernorm.weight": "model-00016-of-00040.safetensors",
45
+ "model.layers.12.mlp.down_proj.weight": "model-00016-of-00040.safetensors",
46
+ "model.layers.12.mlp.gate_proj.weight": "model-00016-of-00040.safetensors",
47
+ "model.layers.12.mlp.up_proj.weight": "model-00016-of-00040.safetensors",
48
+ "model.layers.12.post_attention_layernorm.weight": "model-00016-of-00040.safetensors",
49
+ "model.layers.12.self_attn.k_proj.weight": "model-00016-of-00040.safetensors",
50
+ "model.layers.12.self_attn.o_proj.weight": "model-00016-of-00040.safetensors",
51
+ "model.layers.12.self_attn.q_proj.weight": "model-00015-of-00040.safetensors",
52
+ "model.layers.12.self_attn.v_proj.weight": "model-00016-of-00040.safetensors",
53
+ "model.layers.13.input_layernorm.weight": "model-00018-of-00040.safetensors",
54
+ "model.layers.13.mlp.down_proj.weight": "model-00018-of-00040.safetensors",
55
+ "model.layers.13.mlp.gate_proj.weight": "model-00017-of-00040.safetensors",
56
+ "model.layers.13.mlp.up_proj.weight": "model-00017-of-00040.safetensors",
57
+ "model.layers.13.post_attention_layernorm.weight": "model-00018-of-00040.safetensors",
58
+ "model.layers.13.self_attn.k_proj.weight": "model-00017-of-00040.safetensors",
59
+ "model.layers.13.self_attn.o_proj.weight": "model-00017-of-00040.safetensors",
60
+ "model.layers.13.self_attn.q_proj.weight": "model-00017-of-00040.safetensors",
61
+ "model.layers.13.self_attn.v_proj.weight": "model-00017-of-00040.safetensors",
62
+ "model.layers.14.input_layernorm.weight": "model-00019-of-00040.safetensors",
63
+ "model.layers.14.mlp.down_proj.weight": "model-00019-of-00040.safetensors",
64
+ "model.layers.14.mlp.gate_proj.weight": "model-00018-of-00040.safetensors",
65
+ "model.layers.14.mlp.up_proj.weight": "model-00019-of-00040.safetensors",
66
+ "model.layers.14.post_attention_layernorm.weight": "model-00019-of-00040.safetensors",
67
+ "model.layers.14.self_attn.k_proj.weight": "model-00018-of-00040.safetensors",
68
+ "model.layers.14.self_attn.o_proj.weight": "model-00018-of-00040.safetensors",
69
+ "model.layers.14.self_attn.q_proj.weight": "model-00018-of-00040.safetensors",
70
+ "model.layers.14.self_attn.v_proj.weight": "model-00018-of-00040.safetensors",
71
+ "model.layers.15.input_layernorm.weight": "model-00020-of-00040.safetensors",
72
+ "model.layers.15.mlp.down_proj.weight": "model-00020-of-00040.safetensors",
73
+ "model.layers.15.mlp.gate_proj.weight": "model-00020-of-00040.safetensors",
74
+ "model.layers.15.mlp.up_proj.weight": "model-00020-of-00040.safetensors",
75
+ "model.layers.15.post_attention_layernorm.weight": "model-00020-of-00040.safetensors",
76
+ "model.layers.15.self_attn.k_proj.weight": "model-00019-of-00040.safetensors",
77
+ "model.layers.15.self_attn.o_proj.weight": "model-00019-of-00040.safetensors",
78
+ "model.layers.15.self_attn.q_proj.weight": "model-00019-of-00040.safetensors",
79
+ "model.layers.15.self_attn.v_proj.weight": "model-00019-of-00040.safetensors",
80
+ "model.layers.16.input_layernorm.weight": "model-00021-of-00040.safetensors",
81
+ "model.layers.16.mlp.down_proj.weight": "model-00021-of-00040.safetensors",
82
+ "model.layers.16.mlp.gate_proj.weight": "model-00021-of-00040.safetensors",
83
+ "model.layers.16.mlp.up_proj.weight": "model-00021-of-00040.safetensors",
84
+ "model.layers.16.post_attention_layernorm.weight": "model-00021-of-00040.safetensors",
85
+ "model.layers.16.self_attn.k_proj.weight": "model-00020-of-00040.safetensors",
86
+ "model.layers.16.self_attn.o_proj.weight": "model-00021-of-00040.safetensors",
87
+ "model.layers.16.self_attn.q_proj.weight": "model-00020-of-00040.safetensors",
88
+ "model.layers.16.self_attn.v_proj.weight": "model-00020-of-00040.safetensors",
89
+ "model.layers.17.input_layernorm.weight": "model-00022-of-00040.safetensors",
90
+ "model.layers.17.mlp.down_proj.weight": "model-00022-of-00040.safetensors",
91
+ "model.layers.17.mlp.gate_proj.weight": "model-00022-of-00040.safetensors",
92
+ "model.layers.17.mlp.up_proj.weight": "model-00022-of-00040.safetensors",
93
+ "model.layers.17.post_attention_layernorm.weight": "model-00022-of-00040.safetensors",
94
+ "model.layers.17.self_attn.k_proj.weight": "model-00021-of-00040.safetensors",
95
+ "model.layers.17.self_attn.o_proj.weight": "model-00022-of-00040.safetensors",
96
+ "model.layers.17.self_attn.q_proj.weight": "model-00021-of-00040.safetensors",
97
+ "model.layers.17.self_attn.v_proj.weight": "model-00022-of-00040.safetensors",
98
+ "model.layers.18.input_layernorm.weight": "model-00023-of-00040.safetensors",
99
+ "model.layers.18.mlp.down_proj.weight": "model-00023-of-00040.safetensors",
100
+ "model.layers.18.mlp.gate_proj.weight": "model-00023-of-00040.safetensors",
101
+ "model.layers.18.mlp.up_proj.weight": "model-00023-of-00040.safetensors",
102
+ "model.layers.18.post_attention_layernorm.weight": "model-00023-of-00040.safetensors",
103
+ "model.layers.18.self_attn.k_proj.weight": "model-00023-of-00040.safetensors",
104
+ "model.layers.18.self_attn.o_proj.weight": "model-00023-of-00040.safetensors",
105
+ "model.layers.18.self_attn.q_proj.weight": "model-00022-of-00040.safetensors",
106
+ "model.layers.18.self_attn.v_proj.weight": "model-00023-of-00040.safetensors",
107
+ "model.layers.19.input_layernorm.weight": "model-00025-of-00040.safetensors",
108
+ "model.layers.19.mlp.down_proj.weight": "model-00025-of-00040.safetensors",
109
+ "model.layers.19.mlp.gate_proj.weight": "model-00024-of-00040.safetensors",
110
+ "model.layers.19.mlp.up_proj.weight": "model-00024-of-00040.safetensors",
111
+ "model.layers.19.post_attention_layernorm.weight": "model-00025-of-00040.safetensors",
112
+ "model.layers.19.self_attn.k_proj.weight": "model-00024-of-00040.safetensors",
113
+ "model.layers.19.self_attn.o_proj.weight": "model-00024-of-00040.safetensors",
114
+ "model.layers.19.self_attn.q_proj.weight": "model-00024-of-00040.safetensors",
115
+ "model.layers.19.self_attn.v_proj.weight": "model-00024-of-00040.safetensors",
116
+ "model.layers.2.input_layernorm.weight": "model-00005-of-00040.safetensors",
117
+ "model.layers.2.mlp.down_proj.weight": "model-00005-of-00040.safetensors",
118
+ "model.layers.2.mlp.gate_proj.weight": "model-00004-of-00040.safetensors",
119
+ "model.layers.2.mlp.up_proj.weight": "model-00005-of-00040.safetensors",
120
+ "model.layers.2.post_attention_layernorm.weight": "model-00005-of-00040.safetensors",
121
+ "model.layers.2.self_attn.k_proj.weight": "model-00004-of-00040.safetensors",
122
+ "model.layers.2.self_attn.o_proj.weight": "model-00004-of-00040.safetensors",
123
+ "model.layers.2.self_attn.q_proj.weight": "model-00004-of-00040.safetensors",
124
+ "model.layers.2.self_attn.v_proj.weight": "model-00004-of-00040.safetensors",
125
+ "model.layers.20.input_layernorm.weight": "model-00026-of-00040.safetensors",
126
+ "model.layers.20.mlp.down_proj.weight": "model-00026-of-00040.safetensors",
127
+ "model.layers.20.mlp.gate_proj.weight": "model-00025-of-00040.safetensors",
128
+ "model.layers.20.mlp.up_proj.weight": "model-00026-of-00040.safetensors",
129
+ "model.layers.20.post_attention_layernorm.weight": "model-00026-of-00040.safetensors",
130
+ "model.layers.20.self_attn.k_proj.weight": "model-00025-of-00040.safetensors",
131
+ "model.layers.20.self_attn.o_proj.weight": "model-00025-of-00040.safetensors",
132
+ "model.layers.20.self_attn.q_proj.weight": "model-00025-of-00040.safetensors",
133
+ "model.layers.20.self_attn.v_proj.weight": "model-00025-of-00040.safetensors",
134
+ "model.layers.21.input_layernorm.weight": "model-00027-of-00040.safetensors",
135
+ "model.layers.21.mlp.down_proj.weight": "model-00027-of-00040.safetensors",
136
+ "model.layers.21.mlp.gate_proj.weight": "model-00027-of-00040.safetensors",
137
+ "model.layers.21.mlp.up_proj.weight": "model-00027-of-00040.safetensors",
138
+ "model.layers.21.post_attention_layernorm.weight": "model-00027-of-00040.safetensors",
139
+ "model.layers.21.self_attn.k_proj.weight": "model-00026-of-00040.safetensors",
140
+ "model.layers.21.self_attn.o_proj.weight": "model-00026-of-00040.safetensors",
141
+ "model.layers.21.self_attn.q_proj.weight": "model-00026-of-00040.safetensors",
142
+ "model.layers.21.self_attn.v_proj.weight": "model-00026-of-00040.safetensors",
143
+ "model.layers.22.input_layernorm.weight": "model-00028-of-00040.safetensors",
144
+ "model.layers.22.mlp.down_proj.weight": "model-00028-of-00040.safetensors",
145
+ "model.layers.22.mlp.gate_proj.weight": "model-00028-of-00040.safetensors",
146
+ "model.layers.22.mlp.up_proj.weight": "model-00028-of-00040.safetensors",
147
+ "model.layers.22.post_attention_layernorm.weight": "model-00028-of-00040.safetensors",
148
+ "model.layers.22.self_attn.k_proj.weight": "model-00027-of-00040.safetensors",
149
+ "model.layers.22.self_attn.o_proj.weight": "model-00028-of-00040.safetensors",
150
+ "model.layers.22.self_attn.q_proj.weight": "model-00027-of-00040.safetensors",
151
+ "model.layers.22.self_attn.v_proj.weight": "model-00027-of-00040.safetensors",
152
+ "model.layers.23.input_layernorm.weight": "model-00029-of-00040.safetensors",
153
+ "model.layers.23.mlp.down_proj.weight": "model-00029-of-00040.safetensors",
154
+ "model.layers.23.mlp.gate_proj.weight": "model-00029-of-00040.safetensors",
155
+ "model.layers.23.mlp.up_proj.weight": "model-00029-of-00040.safetensors",
156
+ "model.layers.23.post_attention_layernorm.weight": "model-00029-of-00040.safetensors",
157
+ "model.layers.23.self_attn.k_proj.weight": "model-00028-of-00040.safetensors",
158
+ "model.layers.23.self_attn.o_proj.weight": "model-00029-of-00040.safetensors",
159
+ "model.layers.23.self_attn.q_proj.weight": "model-00028-of-00040.safetensors",
160
+ "model.layers.23.self_attn.v_proj.weight": "model-00029-of-00040.safetensors",
161
+ "model.layers.24.input_layernorm.weight": "model-00030-of-00040.safetensors",
162
+ "model.layers.24.mlp.down_proj.weight": "model-00030-of-00040.safetensors",
163
+ "model.layers.24.mlp.gate_proj.weight": "model-00030-of-00040.safetensors",
164
+ "model.layers.24.mlp.up_proj.weight": "model-00030-of-00040.safetensors",
165
+ "model.layers.24.post_attention_layernorm.weight": "model-00030-of-00040.safetensors",
166
+ "model.layers.24.self_attn.k_proj.weight": "model-00030-of-00040.safetensors",
167
+ "model.layers.24.self_attn.o_proj.weight": "model-00030-of-00040.safetensors",
168
+ "model.layers.24.self_attn.q_proj.weight": "model-00029-of-00040.safetensors",
169
+ "model.layers.24.self_attn.v_proj.weight": "model-00030-of-00040.safetensors",
170
+ "model.layers.25.input_layernorm.weight": "model-00032-of-00040.safetensors",
171
+ "model.layers.25.mlp.down_proj.weight": "model-00032-of-00040.safetensors",
172
+ "model.layers.25.mlp.gate_proj.weight": "model-00031-of-00040.safetensors",
173
+ "model.layers.25.mlp.up_proj.weight": "model-00031-of-00040.safetensors",
174
+ "model.layers.25.post_attention_layernorm.weight": "model-00032-of-00040.safetensors",
175
+ "model.layers.25.self_attn.k_proj.weight": "model-00031-of-00040.safetensors",
176
+ "model.layers.25.self_attn.o_proj.weight": "model-00031-of-00040.safetensors",
177
+ "model.layers.25.self_attn.q_proj.weight": "model-00031-of-00040.safetensors",
178
+ "model.layers.25.self_attn.v_proj.weight": "model-00031-of-00040.safetensors",
179
+ "model.layers.26.input_layernorm.weight": "model-00033-of-00040.safetensors",
180
+ "model.layers.26.mlp.down_proj.weight": "model-00033-of-00040.safetensors",
181
+ "model.layers.26.mlp.gate_proj.weight": "model-00032-of-00040.safetensors",
182
+ "model.layers.26.mlp.up_proj.weight": "model-00033-of-00040.safetensors",
183
+ "model.layers.26.post_attention_layernorm.weight": "model-00033-of-00040.safetensors",
184
+ "model.layers.26.self_attn.k_proj.weight": "model-00032-of-00040.safetensors",
185
+ "model.layers.26.self_attn.o_proj.weight": "model-00032-of-00040.safetensors",
186
+ "model.layers.26.self_attn.q_proj.weight": "model-00032-of-00040.safetensors",
187
+ "model.layers.26.self_attn.v_proj.weight": "model-00032-of-00040.safetensors",
188
+ "model.layers.27.input_layernorm.weight": "model-00034-of-00040.safetensors",
189
+ "model.layers.27.mlp.down_proj.weight": "model-00034-of-00040.safetensors",
190
+ "model.layers.27.mlp.gate_proj.weight": "model-00034-of-00040.safetensors",
191
+ "model.layers.27.mlp.up_proj.weight": "model-00034-of-00040.safetensors",
192
+ "model.layers.27.post_attention_layernorm.weight": "model-00034-of-00040.safetensors",
193
+ "model.layers.27.self_attn.k_proj.weight": "model-00033-of-00040.safetensors",
194
+ "model.layers.27.self_attn.o_proj.weight": "model-00033-of-00040.safetensors",
195
+ "model.layers.27.self_attn.q_proj.weight": "model-00033-of-00040.safetensors",
196
+ "model.layers.27.self_attn.v_proj.weight": "model-00033-of-00040.safetensors",
197
+ "model.layers.28.input_layernorm.weight": "model-00035-of-00040.safetensors",
198
+ "model.layers.28.mlp.down_proj.weight": "model-00035-of-00040.safetensors",
199
+ "model.layers.28.mlp.gate_proj.weight": "model-00035-of-00040.safetensors",
200
+ "model.layers.28.mlp.up_proj.weight": "model-00035-of-00040.safetensors",
201
+ "model.layers.28.post_attention_layernorm.weight": "model-00035-of-00040.safetensors",
202
+ "model.layers.28.self_attn.k_proj.weight": "model-00034-of-00040.safetensors",
203
+ "model.layers.28.self_attn.o_proj.weight": "model-00035-of-00040.safetensors",
204
+ "model.layers.28.self_attn.q_proj.weight": "model-00034-of-00040.safetensors",
205
+ "model.layers.28.self_attn.v_proj.weight": "model-00034-of-00040.safetensors",
206
+ "model.layers.29.input_layernorm.weight": "model-00036-of-00040.safetensors",
207
+ "model.layers.29.mlp.down_proj.weight": "model-00036-of-00040.safetensors",
208
+ "model.layers.29.mlp.gate_proj.weight": "model-00036-of-00040.safetensors",
209
+ "model.layers.29.mlp.up_proj.weight": "model-00036-of-00040.safetensors",
210
+ "model.layers.29.post_attention_layernorm.weight": "model-00036-of-00040.safetensors",
211
+ "model.layers.29.self_attn.k_proj.weight": "model-00035-of-00040.safetensors",
212
+ "model.layers.29.self_attn.o_proj.weight": "model-00036-of-00040.safetensors",
213
+ "model.layers.29.self_attn.q_proj.weight": "model-00035-of-00040.safetensors",
214
+ "model.layers.29.self_attn.v_proj.weight": "model-00036-of-00040.safetensors",
215
+ "model.layers.3.input_layernorm.weight": "model-00006-of-00040.safetensors",
216
+ "model.layers.3.mlp.down_proj.weight": "model-00006-of-00040.safetensors",
217
+ "model.layers.3.mlp.gate_proj.weight": "model-00006-of-00040.safetensors",
218
+ "model.layers.3.mlp.up_proj.weight": "model-00006-of-00040.safetensors",
219
+ "model.layers.3.post_attention_layernorm.weight": "model-00006-of-00040.safetensors",
220
+ "model.layers.3.self_attn.k_proj.weight": "model-00005-of-00040.safetensors",
221
+ "model.layers.3.self_attn.o_proj.weight": "model-00005-of-00040.safetensors",
222
+ "model.layers.3.self_attn.q_proj.weight": "model-00005-of-00040.safetensors",
223
+ "model.layers.3.self_attn.v_proj.weight": "model-00005-of-00040.safetensors",
224
+ "model.layers.30.input_layernorm.weight": "model-00037-of-00040.safetensors",
225
+ "model.layers.30.mlp.down_proj.weight": "model-00037-of-00040.safetensors",
226
+ "model.layers.30.mlp.gate_proj.weight": "model-00037-of-00040.safetensors",
227
+ "model.layers.30.mlp.up_proj.weight": "model-00037-of-00040.safetensors",
228
+ "model.layers.30.post_attention_layernorm.weight": "model-00037-of-00040.safetensors",
229
+ "model.layers.30.self_attn.k_proj.weight": "model-00037-of-00040.safetensors",
230
+ "model.layers.30.self_attn.o_proj.weight": "model-00037-of-00040.safetensors",
231
+ "model.layers.30.self_attn.q_proj.weight": "model-00036-of-00040.safetensors",
232
+ "model.layers.30.self_attn.v_proj.weight": "model-00037-of-00040.safetensors",
233
+ "model.layers.31.input_layernorm.weight": "model-00039-of-00040.safetensors",
234
+ "model.layers.31.mlp.down_proj.weight": "model-00039-of-00040.safetensors",
235
+ "model.layers.31.mlp.gate_proj.weight": "model-00038-of-00040.safetensors",
236
+ "model.layers.31.mlp.up_proj.weight": "model-00038-of-00040.safetensors",
237
+ "model.layers.31.post_attention_layernorm.weight": "model-00039-of-00040.safetensors",
238
+ "model.layers.31.self_attn.k_proj.weight": "model-00038-of-00040.safetensors",
239
+ "model.layers.31.self_attn.o_proj.weight": "model-00038-of-00040.safetensors",
240
+ "model.layers.31.self_attn.q_proj.weight": "model-00038-of-00040.safetensors",
241
+ "model.layers.31.self_attn.v_proj.weight": "model-00038-of-00040.safetensors",
242
+ "model.layers.4.input_layernorm.weight": "model-00007-of-00040.safetensors",
243
+ "model.layers.4.mlp.down_proj.weight": "model-00007-of-00040.safetensors",
244
+ "model.layers.4.mlp.gate_proj.weight": "model-00007-of-00040.safetensors",
245
+ "model.layers.4.mlp.up_proj.weight": "model-00007-of-00040.safetensors",
246
+ "model.layers.4.post_attention_layernorm.weight": "model-00007-of-00040.safetensors",
247
+ "model.layers.4.self_attn.k_proj.weight": "model-00006-of-00040.safetensors",
248
+ "model.layers.4.self_attn.o_proj.weight": "model-00007-of-00040.safetensors",
249
+ "model.layers.4.self_attn.q_proj.weight": "model-00006-of-00040.safetensors",
250
+ "model.layers.4.self_attn.v_proj.weight": "model-00006-of-00040.safetensors",
251
+ "model.layers.5.input_layernorm.weight": "model-00008-of-00040.safetensors",
252
+ "model.layers.5.mlp.down_proj.weight": "model-00008-of-00040.safetensors",
253
+ "model.layers.5.mlp.gate_proj.weight": "model-00008-of-00040.safetensors",
254
+ "model.layers.5.mlp.up_proj.weight": "model-00008-of-00040.safetensors",
255
+ "model.layers.5.post_attention_layernorm.weight": "model-00008-of-00040.safetensors",
256
+ "model.layers.5.self_attn.k_proj.weight": "model-00007-of-00040.safetensors",
257
+ "model.layers.5.self_attn.o_proj.weight": "model-00008-of-00040.safetensors",
258
+ "model.layers.5.self_attn.q_proj.weight": "model-00007-of-00040.safetensors",
259
+ "model.layers.5.self_attn.v_proj.weight": "model-00008-of-00040.safetensors",
260
+ "model.layers.6.input_layernorm.weight": "model-00009-of-00040.safetensors",
261
+ "model.layers.6.mlp.down_proj.weight": "model-00009-of-00040.safetensors",
262
+ "model.layers.6.mlp.gate_proj.weight": "model-00009-of-00040.safetensors",
263
+ "model.layers.6.mlp.up_proj.weight": "model-00009-of-00040.safetensors",
264
+ "model.layers.6.post_attention_layernorm.weight": "model-00009-of-00040.safetensors",
265
+ "model.layers.6.self_attn.k_proj.weight": "model-00009-of-00040.safetensors",
266
+ "model.layers.6.self_attn.o_proj.weight": "model-00009-of-00040.safetensors",
267
+ "model.layers.6.self_attn.q_proj.weight": "model-00008-of-00040.safetensors",
268
+ "model.layers.6.self_attn.v_proj.weight": "model-00009-of-00040.safetensors",
269
+ "model.layers.7.input_layernorm.weight": "model-00011-of-00040.safetensors",
270
+ "model.layers.7.mlp.down_proj.weight": "model-00011-of-00040.safetensors",
271
+ "model.layers.7.mlp.gate_proj.weight": "model-00010-of-00040.safetensors",
272
+ "model.layers.7.mlp.up_proj.weight": "model-00010-of-00040.safetensors",
273
+ "model.layers.7.post_attention_layernorm.weight": "model-00011-of-00040.safetensors",
274
+ "model.layers.7.self_attn.k_proj.weight": "model-00010-of-00040.safetensors",
275
+ "model.layers.7.self_attn.o_proj.weight": "model-00010-of-00040.safetensors",
276
+ "model.layers.7.self_attn.q_proj.weight": "model-00010-of-00040.safetensors",
277
+ "model.layers.7.self_attn.v_proj.weight": "model-00010-of-00040.safetensors",
278
+ "model.layers.8.input_layernorm.weight": "model-00012-of-00040.safetensors",
279
+ "model.layers.8.mlp.down_proj.weight": "model-00012-of-00040.safetensors",
280
+ "model.layers.8.mlp.gate_proj.weight": "model-00011-of-00040.safetensors",
281
+ "model.layers.8.mlp.up_proj.weight": "model-00012-of-00040.safetensors",
282
+ "model.layers.8.post_attention_layernorm.weight": "model-00012-of-00040.safetensors",
283
+ "model.layers.8.self_attn.k_proj.weight": "model-00011-of-00040.safetensors",
284
+ "model.layers.8.self_attn.o_proj.weight": "model-00011-of-00040.safetensors",
285
+ "model.layers.8.self_attn.q_proj.weight": "model-00011-of-00040.safetensors",
286
+ "model.layers.8.self_attn.v_proj.weight": "model-00011-of-00040.safetensors",
287
+ "model.layers.9.input_layernorm.weight": "model-00013-of-00040.safetensors",
288
+ "model.layers.9.mlp.down_proj.weight": "model-00013-of-00040.safetensors",
289
+ "model.layers.9.mlp.gate_proj.weight": "model-00013-of-00040.safetensors",
290
+ "model.layers.9.mlp.up_proj.weight": "model-00013-of-00040.safetensors",
291
+ "model.layers.9.post_attention_layernorm.weight": "model-00013-of-00040.safetensors",
292
+ "model.layers.9.self_attn.k_proj.weight": "model-00012-of-00040.safetensors",
293
+ "model.layers.9.self_attn.o_proj.weight": "model-00012-of-00040.safetensors",
294
+ "model.layers.9.self_attn.q_proj.weight": "model-00012-of-00040.safetensors",
295
+ "model.layers.9.self_attn.v_proj.weight": "model-00012-of-00040.safetensors",
296
+ "model.norm.weight": "model-00039-of-00040.safetensors"
297
+ }
298
+ }