Upload folder using huggingface_hub

config.json

@@ -0,0 +1,30 @@
+{
+  "_name_or_path": "nlpai-lab/kullm-polyglot-12.8b-v2",
+  "architectures": [
+    "GPTNeoXForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classifier_dropout": 0.1,
+  "eos_token_id": 0,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.0,
+  "hidden_size": 5120,
+  "initializer_range": 0.02,
+  "intermediate_size": 20480,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 2048,
+  "model_type": "gpt_neox",
+  "num_attention_heads": 40,
+  "num_hidden_layers": 40,
+  "num_steps": "global_step301000",
+  "rope_scaling": null,
+  "rotary_emb_base": 10000,
+  "rotary_pct": 0.5,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.34.0",
+  "use_cache": true,
+  "use_parallel_residual": true,
+  "vocab_size": 30080
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 0,
+  "eos_token_id": 0,
+  "transformers_version": "4.34.0"
+}

model-00001-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66408da03ee24e97df968ea259787d830e3aeaa2317b52efffd13f519badbdfe
+size 937341328

model-00002-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c90848b77731bb4726cd69a19a14053290aeb92b628818f4b10f6b31436f6cc1
+size 839037000

model-00003-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc74458b83e44293fc35d02cd660a41e7617eba3fc072b120d6b8d2d6056d7ee
+size 839036760

model-00004-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0cb8a2a27c52f8f149ce3e5dc0c97528ed25e4527fc52870a448bc722d12bbf2
+size 996365056

model-00005-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd768bbec3b80dd02bb970d125f360d57863bb200a3ea91d108af2d675515fcf
+size 891476024

model-00006-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:626ee2f2413d83a8bbb829a09d5ec54474db9a59a6b5a89dc6ec2b9aa6fc132c
+size 996365056

model-00007-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:23a2f90759e6ce32b2bf48832a94f54685713c5ac355162ad24e514d4fdbd92f
+size 891476024

model-00008-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cead3a5ed3bb62d1bcca8c844c7fc78f99581d13cbe79d287a9d7a5fadb93bc1
+size 996365072

model-00009-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:549d8ad6fd7a582ead6e76eb9f45e459a577adbf783080f21ed88ec1e43ffe4b
+size 891476040

model-00010-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff96011ef9c061508defe3fa7c341a92566186173809654b5410edc95d906d6a
+size 996365080

model-00011-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27ad543d47a8c212d98e7166ef113b6c8e39f2842505ab95c868873ec8f88f1b
+size 891476040

model-00012-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:547f56abd19d4f0fcfa2296ec0aa57ed1fb5adcf134b9980ce7980b71ebba4dd
+size 996365080

model-00013-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c729dfed10ba8c0d4e628b14945a9f21fe98a2578381743d4adb98a5cde7b7b6
+size 891476040

model-00014-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a2f1c6ac8f2856780a0c89887dc3f5da01c1213adf84055b1128149e7aa4dff
+size 996365080

model-00015-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:408264170ad2a0291d66f515d9e11f4bfa6498a2296c0884bb5dbc1c78763318
+size 891476040

model-00016-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:99c88e1e47f3fd04d45e7d1446135484f2174c768cf5b373e6f809d6f4b8cb7c
+size 996365080

model-00017-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ddafcc2bfd8e9824397e1efc5c7f535c61ab549bf4cd9e896e64e9fce78d057c
+size 891476040

model-00018-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:523252277649a5a159e0b1cd7115fc20164bb406e4bac973c04c853606041ce1
+size 996365080

model-00019-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dcc20034a6ebcdc22c720330d0e04cdd0ecf2835cf81af6777c4ee1d82522140
+size 891476040

model-00020-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c38b05dc4524320978c0a8e5b955e283cb9127d1bd66df6f0db038eb6eec261a
+size 996365080

model-00021-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6153bd7ecfe7355156458ce204d97bbcfa85995b511ff033492641188adbfa32
+size 891476040

model-00022-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c52f5d00cd9c3fdc07ccfc823305d634b623144bc59a83e8900fa8b25aeccc6
+size 996365080

model-00023-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d93a251d30fc4f3ad7bc939d72f9bf044f9ef714be581d8cbde3da58a0580418
+size 891476040

model-00024-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9828e8f97ea5a360f39ded5f90086a6a4d8732649f16d03e099502c63d37f44f
+size 996365080

model-00025-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d82ec37a309bf8542090863f21e75f2bbe9858338e8a8be7f81c2d2559be8037
+size 891476040

model-00026-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76eb3d56798d6f17c9e53c4141b5dae3fbef1d4fb725b410ff94ab2e4ac238c8
+size 996365080

model-00027-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70c8b859e4f07d091480c912e04f0c364647cf4988b775e7e2188ac2fbeed49b
+size 891476040

model-00028-of-00028.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7ea0188f126c4cd893bfdf40fbb5fb222da030f321ce595dfcf855887d72ead
+size 517765688

model.safetensors.index.json

@@ -0,0 +1,531 @@
+{
+  "metadata": {
+    "total_size": 25787212800
+  },
+  "weight_map": {
+    "embed_out.weight": "model-00028-of-00028.safetensors",
+    "gpt_neox.embed_in.weight": "model-00001-of-00028.safetensors",
+    "gpt_neox.final_layer_norm.bias": "model-00028-of-00028.safetensors",
+    "gpt_neox.final_layer_norm.weight": "model-00028-of-00028.safetensors",
+    "gpt_neox.layers.0.attention.dense.bias": "model-00001-of-00028.safetensors",
+    "gpt_neox.layers.0.attention.dense.weight": "model-00001-of-00028.safetensors",
+    "gpt_neox.layers.0.attention.query_key_value.bias": "model-00001-of-00028.safetensors",
+    "gpt_neox.layers.0.attention.query_key_value.weight": "model-00001-of-00028.safetensors",
+    "gpt_neox.layers.0.attention.rotary_emb.inv_freq": "model-00001-of-00028.safetensors",
+    "gpt_neox.layers.0.input_layernorm.bias": "model-00001-of-00028.safetensors",
+    "gpt_neox.layers.0.input_layernorm.weight": "model-00001-of-00028.safetensors",
+    "gpt_neox.layers.0.mlp.dense_4h_to_h.bias": "model-00001-of-00028.safetensors",
+    "gpt_neox.layers.0.mlp.dense_4h_to_h.weight": "model-00001-of-00028.safetensors",
+    "gpt_neox.layers.0.mlp.dense_h_to_4h.bias": "model-00001-of-00028.safetensors",
+    "gpt_neox.layers.0.mlp.dense_h_to_4h.weight": "model-00001-of-00028.safetensors",
+    "gpt_neox.layers.0.post_attention_layernorm.bias": "model-00001-of-00028.safetensors",
+    "gpt_neox.layers.0.post_attention_layernorm.weight": "model-00001-of-00028.safetensors",
+    "gpt_neox.layers.1.attention.dense.bias": "model-00002-of-00028.safetensors",
+    "gpt_neox.layers.1.attention.dense.weight": "model-00002-of-00028.safetensors",
+    "gpt_neox.layers.1.attention.query_key_value.bias": "model-00002-of-00028.safetensors",
+    "gpt_neox.layers.1.attention.query_key_value.weight": "model-00002-of-00028.safetensors",
+    "gpt_neox.layers.1.attention.rotary_emb.inv_freq": "model-00001-of-00028.safetensors",
+    "gpt_neox.layers.1.input_layernorm.bias": "model-00001-of-00028.safetensors",
+    "gpt_neox.layers.1.input_layernorm.weight": "model-00001-of-00028.safetensors",
+    "gpt_neox.layers.1.mlp.dense_4h_to_h.bias": "model-00002-of-00028.safetensors",
+    "gpt_neox.layers.1.mlp.dense_4h_to_h.weight": "model-00002-of-00028.safetensors",
+    "gpt_neox.layers.1.mlp.dense_h_to_4h.bias": "model-00002-of-00028.safetensors",
+    "gpt_neox.layers.1.mlp.dense_h_to_4h.weight": "model-00002-of-00028.safetensors",
+    "gpt_neox.layers.1.post_attention_layernorm.bias": "model-00001-of-00028.safetensors",
+    "gpt_neox.layers.1.post_attention_layernorm.weight": "model-00001-of-00028.safetensors",
+    "gpt_neox.layers.10.attention.dense.bias": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.10.attention.dense.weight": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.10.attention.query_key_value.bias": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.10.attention.query_key_value.weight": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.10.attention.rotary_emb.inv_freq": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.10.input_layernorm.bias": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.10.input_layernorm.weight": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.10.mlp.dense_4h_to_h.bias": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.10.mlp.dense_4h_to_h.weight": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.10.mlp.dense_h_to_4h.bias": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.10.mlp.dense_h_to_4h.weight": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.10.post_attention_layernorm.bias": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.10.post_attention_layernorm.weight": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.11.attention.dense.bias": "model-00009-of-00028.safetensors",
+    "gpt_neox.layers.11.attention.dense.weight": "model-00009-of-00028.safetensors",
+    "gpt_neox.layers.11.attention.query_key_value.bias": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.11.attention.query_key_value.weight": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.11.attention.rotary_emb.inv_freq": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.11.input_layernorm.bias": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.11.input_layernorm.weight": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.11.mlp.dense_4h_to_h.bias": "model-00009-of-00028.safetensors",
+    "gpt_neox.layers.11.mlp.dense_4h_to_h.weight": "model-00009-of-00028.safetensors",
+    "gpt_neox.layers.11.mlp.dense_h_to_4h.bias": "model-00009-of-00028.safetensors",
+    "gpt_neox.layers.11.mlp.dense_h_to_4h.weight": "model-00009-of-00028.safetensors",
+    "gpt_neox.layers.11.post_attention_layernorm.bias": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.11.post_attention_layernorm.weight": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.12.attention.dense.bias": "model-00009-of-00028.safetensors",
+    "gpt_neox.layers.12.attention.dense.weight": "model-00009-of-00028.safetensors",
+    "gpt_neox.layers.12.attention.query_key_value.bias": "model-00009-of-00028.safetensors",
+    "gpt_neox.layers.12.attention.query_key_value.weight": "model-00009-of-00028.safetensors",
+    "gpt_neox.layers.12.attention.rotary_emb.inv_freq": "model-00009-of-00028.safetensors",
+    "gpt_neox.layers.12.input_layernorm.bias": "model-00009-of-00028.safetensors",
+    "gpt_neox.layers.12.input_layernorm.weight": "model-00009-of-00028.safetensors",
+    "gpt_neox.layers.12.mlp.dense_4h_to_h.bias": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.12.mlp.dense_4h_to_h.weight": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.12.mlp.dense_h_to_4h.bias": "model-00009-of-00028.safetensors",
+    "gpt_neox.layers.12.mlp.dense_h_to_4h.weight": "model-00009-of-00028.safetensors",
+    "gpt_neox.layers.12.post_attention_layernorm.bias": "model-00009-of-00028.safetensors",
+    "gpt_neox.layers.12.post_attention_layernorm.weight": "model-00009-of-00028.safetensors",
+    "gpt_neox.layers.13.attention.dense.bias": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.13.attention.dense.weight": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.13.attention.query_key_value.bias": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.13.attention.query_key_value.weight": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.13.attention.rotary_emb.inv_freq": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.13.input_layernorm.bias": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.13.input_layernorm.weight": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.13.mlp.dense_4h_to_h.bias": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.13.mlp.dense_4h_to_h.weight": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.13.mlp.dense_h_to_4h.bias": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.13.mlp.dense_h_to_4h.weight": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.13.post_attention_layernorm.bias": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.13.post_attention_layernorm.weight": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.14.attention.dense.bias": "model-00011-of-00028.safetensors",
+    "gpt_neox.layers.14.attention.dense.weight": "model-00011-of-00028.safetensors",
+    "gpt_neox.layers.14.attention.query_key_value.bias": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.14.attention.query_key_value.weight": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.14.attention.rotary_emb.inv_freq": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.14.input_layernorm.bias": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.14.input_layernorm.weight": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.14.mlp.dense_4h_to_h.bias": "model-00011-of-00028.safetensors",
+    "gpt_neox.layers.14.mlp.dense_4h_to_h.weight": "model-00011-of-00028.safetensors",
+    "gpt_neox.layers.14.mlp.dense_h_to_4h.bias": "model-00011-of-00028.safetensors",
+    "gpt_neox.layers.14.mlp.dense_h_to_4h.weight": "model-00011-of-00028.safetensors",
+    "gpt_neox.layers.14.post_attention_layernorm.bias": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.14.post_attention_layernorm.weight": "model-00010-of-00028.safetensors",
+    "gpt_neox.layers.15.attention.dense.bias": "model-00011-of-00028.safetensors",
+    "gpt_neox.layers.15.attention.dense.weight": "model-00011-of-00028.safetensors",
+    "gpt_neox.layers.15.attention.query_key_value.bias": "model-00011-of-00028.safetensors",
+    "gpt_neox.layers.15.attention.query_key_value.weight": "model-00011-of-00028.safetensors",
+    "gpt_neox.layers.15.attention.rotary_emb.inv_freq": "model-00011-of-00028.safetensors",
+    "gpt_neox.layers.15.input_layernorm.bias": "model-00011-of-00028.safetensors",
+    "gpt_neox.layers.15.input_layernorm.weight": "model-00011-of-00028.safetensors",
+    "gpt_neox.layers.15.mlp.dense_4h_to_h.bias": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.15.mlp.dense_4h_to_h.weight": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.15.mlp.dense_h_to_4h.bias": "model-00011-of-00028.safetensors",
+    "gpt_neox.layers.15.mlp.dense_h_to_4h.weight": "model-00011-of-00028.safetensors",
+    "gpt_neox.layers.15.post_attention_layernorm.bias": "model-00011-of-00028.safetensors",
+    "gpt_neox.layers.15.post_attention_layernorm.weight": "model-00011-of-00028.safetensors",
+    "gpt_neox.layers.16.attention.dense.bias": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.16.attention.dense.weight": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.16.attention.query_key_value.bias": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.16.attention.query_key_value.weight": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.16.attention.rotary_emb.inv_freq": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.16.input_layernorm.bias": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.16.input_layernorm.weight": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.16.mlp.dense_4h_to_h.bias": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.16.mlp.dense_4h_to_h.weight": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.16.mlp.dense_h_to_4h.bias": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.16.mlp.dense_h_to_4h.weight": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.16.post_attention_layernorm.bias": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.16.post_attention_layernorm.weight": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.17.attention.dense.bias": "model-00013-of-00028.safetensors",
+    "gpt_neox.layers.17.attention.dense.weight": "model-00013-of-00028.safetensors",
+    "gpt_neox.layers.17.attention.query_key_value.bias": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.17.attention.query_key_value.weight": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.17.attention.rotary_emb.inv_freq": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.17.input_layernorm.bias": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.17.input_layernorm.weight": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.17.mlp.dense_4h_to_h.bias": "model-00013-of-00028.safetensors",
+    "gpt_neox.layers.17.mlp.dense_4h_to_h.weight": "model-00013-of-00028.safetensors",
+    "gpt_neox.layers.17.mlp.dense_h_to_4h.bias": "model-00013-of-00028.safetensors",
+    "gpt_neox.layers.17.mlp.dense_h_to_4h.weight": "model-00013-of-00028.safetensors",
+    "gpt_neox.layers.17.post_attention_layernorm.bias": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.17.post_attention_layernorm.weight": "model-00012-of-00028.safetensors",
+    "gpt_neox.layers.18.attention.dense.bias": "model-00013-of-00028.safetensors",
+    "gpt_neox.layers.18.attention.dense.weight": "model-00013-of-00028.safetensors",
+    "gpt_neox.layers.18.attention.query_key_value.bias": "model-00013-of-00028.safetensors",
+    "gpt_neox.layers.18.attention.query_key_value.weight": "model-00013-of-00028.safetensors",
+    "gpt_neox.layers.18.attention.rotary_emb.inv_freq": "model-00013-of-00028.safetensors",
+    "gpt_neox.layers.18.input_layernorm.bias": "model-00013-of-00028.safetensors",
+    "gpt_neox.layers.18.input_layernorm.weight": "model-00013-of-00028.safetensors",
+    "gpt_neox.layers.18.mlp.dense_4h_to_h.bias": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.18.mlp.dense_4h_to_h.weight": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.18.mlp.dense_h_to_4h.bias": "model-00013-of-00028.safetensors",
+    "gpt_neox.layers.18.mlp.dense_h_to_4h.weight": "model-00013-of-00028.safetensors",
+    "gpt_neox.layers.18.post_attention_layernorm.bias": "model-00013-of-00028.safetensors",
+    "gpt_neox.layers.18.post_attention_layernorm.weight": "model-00013-of-00028.safetensors",
+    "gpt_neox.layers.19.attention.dense.bias": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.19.attention.dense.weight": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.19.attention.query_key_value.bias": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.19.attention.query_key_value.weight": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.19.attention.rotary_emb.inv_freq": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.19.input_layernorm.bias": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.19.input_layernorm.weight": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.19.mlp.dense_4h_to_h.bias": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.19.mlp.dense_4h_to_h.weight": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.19.mlp.dense_h_to_4h.bias": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.19.mlp.dense_h_to_4h.weight": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.19.post_attention_layernorm.bias": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.19.post_attention_layernorm.weight": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.2.attention.dense.bias": "model-00002-of-00028.safetensors",
+    "gpt_neox.layers.2.attention.dense.weight": "model-00002-of-00028.safetensors",
+    "gpt_neox.layers.2.attention.query_key_value.bias": "model-00002-of-00028.safetensors",
+    "gpt_neox.layers.2.attention.query_key_value.weight": "model-00002-of-00028.safetensors",
+    "gpt_neox.layers.2.attention.rotary_emb.inv_freq": "model-00002-of-00028.safetensors",
+    "gpt_neox.layers.2.input_layernorm.bias": "model-00002-of-00028.safetensors",
+    "gpt_neox.layers.2.input_layernorm.weight": "model-00002-of-00028.safetensors",
+    "gpt_neox.layers.2.mlp.dense_4h_to_h.bias": "model-00003-of-00028.safetensors",
+    "gpt_neox.layers.2.mlp.dense_4h_to_h.weight": "model-00003-of-00028.safetensors",
+    "gpt_neox.layers.2.mlp.dense_h_to_4h.bias": "model-00003-of-00028.safetensors",
+    "gpt_neox.layers.2.mlp.dense_h_to_4h.weight": "model-00003-of-00028.safetensors",
+    "gpt_neox.layers.2.post_attention_layernorm.bias": "model-00002-of-00028.safetensors",
+    "gpt_neox.layers.2.post_attention_layernorm.weight": "model-00002-of-00028.safetensors",
+    "gpt_neox.layers.20.attention.dense.bias": "model-00015-of-00028.safetensors",
+    "gpt_neox.layers.20.attention.dense.weight": "model-00015-of-00028.safetensors",
+    "gpt_neox.layers.20.attention.query_key_value.bias": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.20.attention.query_key_value.weight": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.20.attention.rotary_emb.inv_freq": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.20.input_layernorm.bias": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.20.input_layernorm.weight": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.20.mlp.dense_4h_to_h.bias": "model-00015-of-00028.safetensors",
+    "gpt_neox.layers.20.mlp.dense_4h_to_h.weight": "model-00015-of-00028.safetensors",
+    "gpt_neox.layers.20.mlp.dense_h_to_4h.bias": "model-00015-of-00028.safetensors",
+    "gpt_neox.layers.20.mlp.dense_h_to_4h.weight": "model-00015-of-00028.safetensors",
+    "gpt_neox.layers.20.post_attention_layernorm.bias": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.20.post_attention_layernorm.weight": "model-00014-of-00028.safetensors",
+    "gpt_neox.layers.21.attention.dense.bias": "model-00015-of-00028.safetensors",
+    "gpt_neox.layers.21.attention.dense.weight": "model-00015-of-00028.safetensors",
+    "gpt_neox.layers.21.attention.query_key_value.bias": "model-00015-of-00028.safetensors",
+    "gpt_neox.layers.21.attention.query_key_value.weight": "model-00015-of-00028.safetensors",
+    "gpt_neox.layers.21.attention.rotary_emb.inv_freq": "model-00015-of-00028.safetensors",
+    "gpt_neox.layers.21.input_layernorm.bias": "model-00015-of-00028.safetensors",
+    "gpt_neox.layers.21.input_layernorm.weight": "model-00015-of-00028.safetensors",
+    "gpt_neox.layers.21.mlp.dense_4h_to_h.bias": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.21.mlp.dense_4h_to_h.weight": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.21.mlp.dense_h_to_4h.bias": "model-00015-of-00028.safetensors",
+    "gpt_neox.layers.21.mlp.dense_h_to_4h.weight": "model-00015-of-00028.safetensors",
+    "gpt_neox.layers.21.post_attention_layernorm.bias": "model-00015-of-00028.safetensors",
+    "gpt_neox.layers.21.post_attention_layernorm.weight": "model-00015-of-00028.safetensors",
+    "gpt_neox.layers.22.attention.dense.bias": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.22.attention.dense.weight": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.22.attention.query_key_value.bias": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.22.attention.query_key_value.weight": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.22.attention.rotary_emb.inv_freq": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.22.input_layernorm.bias": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.22.input_layernorm.weight": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.22.mlp.dense_4h_to_h.bias": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.22.mlp.dense_4h_to_h.weight": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.22.mlp.dense_h_to_4h.bias": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.22.mlp.dense_h_to_4h.weight": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.22.post_attention_layernorm.bias": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.22.post_attention_layernorm.weight": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.23.attention.dense.bias": "model-00017-of-00028.safetensors",
+    "gpt_neox.layers.23.attention.dense.weight": "model-00017-of-00028.safetensors",
+    "gpt_neox.layers.23.attention.query_key_value.bias": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.23.attention.query_key_value.weight": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.23.attention.rotary_emb.inv_freq": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.23.input_layernorm.bias": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.23.input_layernorm.weight": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.23.mlp.dense_4h_to_h.bias": "model-00017-of-00028.safetensors",
+    "gpt_neox.layers.23.mlp.dense_4h_to_h.weight": "model-00017-of-00028.safetensors",
+    "gpt_neox.layers.23.mlp.dense_h_to_4h.bias": "model-00017-of-00028.safetensors",
+    "gpt_neox.layers.23.mlp.dense_h_to_4h.weight": "model-00017-of-00028.safetensors",
+    "gpt_neox.layers.23.post_attention_layernorm.bias": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.23.post_attention_layernorm.weight": "model-00016-of-00028.safetensors",
+    "gpt_neox.layers.24.attention.dense.bias": "model-00017-of-00028.safetensors",
+    "gpt_neox.layers.24.attention.dense.weight": "model-00017-of-00028.safetensors",
+    "gpt_neox.layers.24.attention.query_key_value.bias": "model-00017-of-00028.safetensors",
+    "gpt_neox.layers.24.attention.query_key_value.weight": "model-00017-of-00028.safetensors",
+    "gpt_neox.layers.24.attention.rotary_emb.inv_freq": "model-00017-of-00028.safetensors",
+    "gpt_neox.layers.24.input_layernorm.bias": "model-00017-of-00028.safetensors",
+    "gpt_neox.layers.24.input_layernorm.weight": "model-00017-of-00028.safetensors",
+    "gpt_neox.layers.24.mlp.dense_4h_to_h.bias": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.24.mlp.dense_4h_to_h.weight": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.24.mlp.dense_h_to_4h.bias": "model-00017-of-00028.safetensors",
+    "gpt_neox.layers.24.mlp.dense_h_to_4h.weight": "model-00017-of-00028.safetensors",
+    "gpt_neox.layers.24.post_attention_layernorm.bias": "model-00017-of-00028.safetensors",
+    "gpt_neox.layers.24.post_attention_layernorm.weight": "model-00017-of-00028.safetensors",
+    "gpt_neox.layers.25.attention.dense.bias": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.25.attention.dense.weight": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.25.attention.query_key_value.bias": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.25.attention.query_key_value.weight": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.25.attention.rotary_emb.inv_freq": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.25.input_layernorm.bias": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.25.input_layernorm.weight": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.25.mlp.dense_4h_to_h.bias": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.25.mlp.dense_4h_to_h.weight": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.25.mlp.dense_h_to_4h.bias": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.25.mlp.dense_h_to_4h.weight": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.25.post_attention_layernorm.bias": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.25.post_attention_layernorm.weight": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.26.attention.dense.bias": "model-00019-of-00028.safetensors",
+    "gpt_neox.layers.26.attention.dense.weight": "model-00019-of-00028.safetensors",
+    "gpt_neox.layers.26.attention.query_key_value.bias": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.26.attention.query_key_value.weight": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.26.attention.rotary_emb.inv_freq": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.26.input_layernorm.bias": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.26.input_layernorm.weight": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.26.mlp.dense_4h_to_h.bias": "model-00019-of-00028.safetensors",
+    "gpt_neox.layers.26.mlp.dense_4h_to_h.weight": "model-00019-of-00028.safetensors",
+    "gpt_neox.layers.26.mlp.dense_h_to_4h.bias": "model-00019-of-00028.safetensors",
+    "gpt_neox.layers.26.mlp.dense_h_to_4h.weight": "model-00019-of-00028.safetensors",
+    "gpt_neox.layers.26.post_attention_layernorm.bias": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.26.post_attention_layernorm.weight": "model-00018-of-00028.safetensors",
+    "gpt_neox.layers.27.attention.dense.bias": "model-00019-of-00028.safetensors",
+    "gpt_neox.layers.27.attention.dense.weight": "model-00019-of-00028.safetensors",
+    "gpt_neox.layers.27.attention.query_key_value.bias": "model-00019-of-00028.safetensors",
+    "gpt_neox.layers.27.attention.query_key_value.weight": "model-00019-of-00028.safetensors",
+    "gpt_neox.layers.27.attention.rotary_emb.inv_freq": "model-00019-of-00028.safetensors",
+    "gpt_neox.layers.27.input_layernorm.bias": "model-00019-of-00028.safetensors",
+    "gpt_neox.layers.27.input_layernorm.weight": "model-00019-of-00028.safetensors",
+    "gpt_neox.layers.27.mlp.dense_4h_to_h.bias": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.27.mlp.dense_4h_to_h.weight": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.27.mlp.dense_h_to_4h.bias": "model-00019-of-00028.safetensors",
+    "gpt_neox.layers.27.mlp.dense_h_to_4h.weight": "model-00019-of-00028.safetensors",
+    "gpt_neox.layers.27.post_attention_layernorm.bias": "model-00019-of-00028.safetensors",
+    "gpt_neox.layers.27.post_attention_layernorm.weight": "model-00019-of-00028.safetensors",
+    "gpt_neox.layers.28.attention.dense.bias": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.28.attention.dense.weight": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.28.attention.query_key_value.bias": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.28.attention.query_key_value.weight": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.28.attention.rotary_emb.inv_freq": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.28.input_layernorm.bias": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.28.input_layernorm.weight": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.28.mlp.dense_4h_to_h.bias": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.28.mlp.dense_4h_to_h.weight": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.28.mlp.dense_h_to_4h.bias": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.28.mlp.dense_h_to_4h.weight": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.28.post_attention_layernorm.bias": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.28.post_attention_layernorm.weight": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.29.attention.dense.bias": "model-00021-of-00028.safetensors",
+    "gpt_neox.layers.29.attention.dense.weight": "model-00021-of-00028.safetensors",
+    "gpt_neox.layers.29.attention.query_key_value.bias": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.29.attention.query_key_value.weight": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.29.attention.rotary_emb.inv_freq": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.29.input_layernorm.bias": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.29.input_layernorm.weight": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.29.mlp.dense_4h_to_h.bias": "model-00021-of-00028.safetensors",
+    "gpt_neox.layers.29.mlp.dense_4h_to_h.weight": "model-00021-of-00028.safetensors",
+    "gpt_neox.layers.29.mlp.dense_h_to_4h.bias": "model-00021-of-00028.safetensors",
+    "gpt_neox.layers.29.mlp.dense_h_to_4h.weight": "model-00021-of-00028.safetensors",
+    "gpt_neox.layers.29.post_attention_layernorm.bias": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.29.post_attention_layernorm.weight": "model-00020-of-00028.safetensors",
+    "gpt_neox.layers.3.attention.dense.bias": "model-00003-of-00028.safetensors",
+    "gpt_neox.layers.3.attention.dense.weight": "model-00003-of-00028.safetensors",
+    "gpt_neox.layers.3.attention.query_key_value.bias": "model-00003-of-00028.safetensors",
+    "gpt_neox.layers.3.attention.query_key_value.weight": "model-00003-of-00028.safetensors",
+    "gpt_neox.layers.3.attention.rotary_emb.inv_freq": "model-00003-of-00028.safetensors",
+    "gpt_neox.layers.3.input_layernorm.bias": "model-00003-of-00028.safetensors",
+    "gpt_neox.layers.3.input_layernorm.weight": "model-00003-of-00028.safetensors",
+    "gpt_neox.layers.3.mlp.dense_4h_to_h.bias": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.3.mlp.dense_4h_to_h.weight": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.3.mlp.dense_h_to_4h.bias": "model-00003-of-00028.safetensors",
+    "gpt_neox.layers.3.mlp.dense_h_to_4h.weight": "model-00003-of-00028.safetensors",
+    "gpt_neox.layers.3.post_attention_layernorm.bias": "model-00003-of-00028.safetensors",
+    "gpt_neox.layers.3.post_attention_layernorm.weight": "model-00003-of-00028.safetensors",
+    "gpt_neox.layers.30.attention.dense.bias": "model-00021-of-00028.safetensors",
+    "gpt_neox.layers.30.attention.dense.weight": "model-00021-of-00028.safetensors",
+    "gpt_neox.layers.30.attention.query_key_value.bias": "model-00021-of-00028.safetensors",
+    "gpt_neox.layers.30.attention.query_key_value.weight": "model-00021-of-00028.safetensors",
+    "gpt_neox.layers.30.attention.rotary_emb.inv_freq": "model-00021-of-00028.safetensors",
+    "gpt_neox.layers.30.input_layernorm.bias": "model-00021-of-00028.safetensors",
+    "gpt_neox.layers.30.input_layernorm.weight": "model-00021-of-00028.safetensors",
+    "gpt_neox.layers.30.mlp.dense_4h_to_h.bias": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.30.mlp.dense_4h_to_h.weight": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.30.mlp.dense_h_to_4h.bias": "model-00021-of-00028.safetensors",
+    "gpt_neox.layers.30.mlp.dense_h_to_4h.weight": "model-00021-of-00028.safetensors",
+    "gpt_neox.layers.30.post_attention_layernorm.bias": "model-00021-of-00028.safetensors",
+    "gpt_neox.layers.30.post_attention_layernorm.weight": "model-00021-of-00028.safetensors",
+    "gpt_neox.layers.31.attention.dense.bias": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.31.attention.dense.weight": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.31.attention.query_key_value.bias": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.31.attention.query_key_value.weight": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.31.attention.rotary_emb.inv_freq": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.31.input_layernorm.bias": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.31.input_layernorm.weight": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.31.mlp.dense_4h_to_h.bias": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.31.mlp.dense_4h_to_h.weight": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.31.mlp.dense_h_to_4h.bias": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.31.mlp.dense_h_to_4h.weight": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.31.post_attention_layernorm.bias": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.31.post_attention_layernorm.weight": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.32.attention.dense.bias": "model-00023-of-00028.safetensors",
+    "gpt_neox.layers.32.attention.dense.weight": "model-00023-of-00028.safetensors",
+    "gpt_neox.layers.32.attention.query_key_value.bias": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.32.attention.query_key_value.weight": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.32.attention.rotary_emb.inv_freq": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.32.input_layernorm.bias": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.32.input_layernorm.weight": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.32.mlp.dense_4h_to_h.bias": "model-00023-of-00028.safetensors",
+    "gpt_neox.layers.32.mlp.dense_4h_to_h.weight": "model-00023-of-00028.safetensors",
+    "gpt_neox.layers.32.mlp.dense_h_to_4h.bias": "model-00023-of-00028.safetensors",
+    "gpt_neox.layers.32.mlp.dense_h_to_4h.weight": "model-00023-of-00028.safetensors",
+    "gpt_neox.layers.32.post_attention_layernorm.bias": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.32.post_attention_layernorm.weight": "model-00022-of-00028.safetensors",
+    "gpt_neox.layers.33.attention.dense.bias": "model-00023-of-00028.safetensors",
+    "gpt_neox.layers.33.attention.dense.weight": "model-00023-of-00028.safetensors",
+    "gpt_neox.layers.33.attention.query_key_value.bias": "model-00023-of-00028.safetensors",
+    "gpt_neox.layers.33.attention.query_key_value.weight": "model-00023-of-00028.safetensors",
+    "gpt_neox.layers.33.attention.rotary_emb.inv_freq": "model-00023-of-00028.safetensors",
+    "gpt_neox.layers.33.input_layernorm.bias": "model-00023-of-00028.safetensors",
+    "gpt_neox.layers.33.input_layernorm.weight": "model-00023-of-00028.safetensors",
+    "gpt_neox.layers.33.mlp.dense_4h_to_h.bias": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.33.mlp.dense_4h_to_h.weight": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.33.mlp.dense_h_to_4h.bias": "model-00023-of-00028.safetensors",
+    "gpt_neox.layers.33.mlp.dense_h_to_4h.weight": "model-00023-of-00028.safetensors",
+    "gpt_neox.layers.33.post_attention_layernorm.bias": "model-00023-of-00028.safetensors",
+    "gpt_neox.layers.33.post_attention_layernorm.weight": "model-00023-of-00028.safetensors",
+    "gpt_neox.layers.34.attention.dense.bias": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.34.attention.dense.weight": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.34.attention.query_key_value.bias": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.34.attention.query_key_value.weight": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.34.attention.rotary_emb.inv_freq": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.34.input_layernorm.bias": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.34.input_layernorm.weight": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.34.mlp.dense_4h_to_h.bias": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.34.mlp.dense_4h_to_h.weight": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.34.mlp.dense_h_to_4h.bias": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.34.mlp.dense_h_to_4h.weight": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.34.post_attention_layernorm.bias": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.34.post_attention_layernorm.weight": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.35.attention.dense.bias": "model-00025-of-00028.safetensors",
+    "gpt_neox.layers.35.attention.dense.weight": "model-00025-of-00028.safetensors",
+    "gpt_neox.layers.35.attention.query_key_value.bias": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.35.attention.query_key_value.weight": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.35.attention.rotary_emb.inv_freq": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.35.input_layernorm.bias": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.35.input_layernorm.weight": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.35.mlp.dense_4h_to_h.bias": "model-00025-of-00028.safetensors",
+    "gpt_neox.layers.35.mlp.dense_4h_to_h.weight": "model-00025-of-00028.safetensors",
+    "gpt_neox.layers.35.mlp.dense_h_to_4h.bias": "model-00025-of-00028.safetensors",
+    "gpt_neox.layers.35.mlp.dense_h_to_4h.weight": "model-00025-of-00028.safetensors",
+    "gpt_neox.layers.35.post_attention_layernorm.bias": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.35.post_attention_layernorm.weight": "model-00024-of-00028.safetensors",
+    "gpt_neox.layers.36.attention.dense.bias": "model-00025-of-00028.safetensors",
+    "gpt_neox.layers.36.attention.dense.weight": "model-00025-of-00028.safetensors",
+    "gpt_neox.layers.36.attention.query_key_value.bias": "model-00025-of-00028.safetensors",
+    "gpt_neox.layers.36.attention.query_key_value.weight": "model-00025-of-00028.safetensors",
+    "gpt_neox.layers.36.attention.rotary_emb.inv_freq": "model-00025-of-00028.safetensors",
+    "gpt_neox.layers.36.input_layernorm.bias": "model-00025-of-00028.safetensors",
+    "gpt_neox.layers.36.input_layernorm.weight": "model-00025-of-00028.safetensors",
+    "gpt_neox.layers.36.mlp.dense_4h_to_h.bias": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.36.mlp.dense_4h_to_h.weight": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.36.mlp.dense_h_to_4h.bias": "model-00025-of-00028.safetensors",
+    "gpt_neox.layers.36.mlp.dense_h_to_4h.weight": "model-00025-of-00028.safetensors",
+    "gpt_neox.layers.36.post_attention_layernorm.bias": "model-00025-of-00028.safetensors",
+    "gpt_neox.layers.36.post_attention_layernorm.weight": "model-00025-of-00028.safetensors",
+    "gpt_neox.layers.37.attention.dense.bias": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.37.attention.dense.weight": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.37.attention.query_key_value.bias": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.37.attention.query_key_value.weight": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.37.attention.rotary_emb.inv_freq": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.37.input_layernorm.bias": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.37.input_layernorm.weight": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.37.mlp.dense_4h_to_h.bias": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.37.mlp.dense_4h_to_h.weight": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.37.mlp.dense_h_to_4h.bias": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.37.mlp.dense_h_to_4h.weight": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.37.post_attention_layernorm.bias": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.37.post_attention_layernorm.weight": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.38.attention.dense.bias": "model-00027-of-00028.safetensors",
+    "gpt_neox.layers.38.attention.dense.weight": "model-00027-of-00028.safetensors",
+    "gpt_neox.layers.38.attention.query_key_value.bias": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.38.attention.query_key_value.weight": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.38.attention.rotary_emb.inv_freq": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.38.input_layernorm.bias": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.38.input_layernorm.weight": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.38.mlp.dense_4h_to_h.bias": "model-00027-of-00028.safetensors",
+    "gpt_neox.layers.38.mlp.dense_4h_to_h.weight": "model-00027-of-00028.safetensors",
+    "gpt_neox.layers.38.mlp.dense_h_to_4h.bias": "model-00027-of-00028.safetensors",
+    "gpt_neox.layers.38.mlp.dense_h_to_4h.weight": "model-00027-of-00028.safetensors",
+    "gpt_neox.layers.38.post_attention_layernorm.bias": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.38.post_attention_layernorm.weight": "model-00026-of-00028.safetensors",
+    "gpt_neox.layers.39.attention.dense.bias": "model-00027-of-00028.safetensors",
+    "gpt_neox.layers.39.attention.dense.weight": "model-00027-of-00028.safetensors",
+    "gpt_neox.layers.39.attention.query_key_value.bias": "model-00027-of-00028.safetensors",
+    "gpt_neox.layers.39.attention.query_key_value.weight": "model-00027-of-00028.safetensors",
+    "gpt_neox.layers.39.attention.rotary_emb.inv_freq": "model-00027-of-00028.safetensors",
+    "gpt_neox.layers.39.input_layernorm.bias": "model-00027-of-00028.safetensors",
+    "gpt_neox.layers.39.input_layernorm.weight": "model-00027-of-00028.safetensors",
+    "gpt_neox.layers.39.mlp.dense_4h_to_h.bias": "model-00028-of-00028.safetensors",
+    "gpt_neox.layers.39.mlp.dense_4h_to_h.weight": "model-00028-of-00028.safetensors",
+    "gpt_neox.layers.39.mlp.dense_h_to_4h.bias": "model-00027-of-00028.safetensors",
+    "gpt_neox.layers.39.mlp.dense_h_to_4h.weight": "model-00027-of-00028.safetensors",
+    "gpt_neox.layers.39.post_attention_layernorm.bias": "model-00027-of-00028.safetensors",
+    "gpt_neox.layers.39.post_attention_layernorm.weight": "model-00027-of-00028.safetensors",
+    "gpt_neox.layers.4.attention.dense.bias": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.4.attention.dense.weight": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.4.attention.query_key_value.bias": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.4.attention.query_key_value.weight": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.4.attention.rotary_emb.inv_freq": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.4.input_layernorm.bias": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.4.input_layernorm.weight": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.4.mlp.dense_4h_to_h.bias": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.4.mlp.dense_4h_to_h.weight": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.4.mlp.dense_h_to_4h.bias": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.4.mlp.dense_h_to_4h.weight": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.4.post_attention_layernorm.bias": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.4.post_attention_layernorm.weight": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.5.attention.dense.bias": "model-00005-of-00028.safetensors",
+    "gpt_neox.layers.5.attention.dense.weight": "model-00005-of-00028.safetensors",
+    "gpt_neox.layers.5.attention.query_key_value.bias": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.5.attention.query_key_value.weight": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.5.attention.rotary_emb.inv_freq": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.5.input_layernorm.bias": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.5.input_layernorm.weight": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.5.mlp.dense_4h_to_h.bias": "model-00005-of-00028.safetensors",
+    "gpt_neox.layers.5.mlp.dense_4h_to_h.weight": "model-00005-of-00028.safetensors",
+    "gpt_neox.layers.5.mlp.dense_h_to_4h.bias": "model-00005-of-00028.safetensors",
+    "gpt_neox.layers.5.mlp.dense_h_to_4h.weight": "model-00005-of-00028.safetensors",
+    "gpt_neox.layers.5.post_attention_layernorm.bias": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.5.post_attention_layernorm.weight": "model-00004-of-00028.safetensors",
+    "gpt_neox.layers.6.attention.dense.bias": "model-00005-of-00028.safetensors",
+    "gpt_neox.layers.6.attention.dense.weight": "model-00005-of-00028.safetensors",
+    "gpt_neox.layers.6.attention.query_key_value.bias": "model-00005-of-00028.safetensors",
+    "gpt_neox.layers.6.attention.query_key_value.weight": "model-00005-of-00028.safetensors",
+    "gpt_neox.layers.6.attention.rotary_emb.inv_freq": "model-00005-of-00028.safetensors",
+    "gpt_neox.layers.6.input_layernorm.bias": "model-00005-of-00028.safetensors",
+    "gpt_neox.layers.6.input_layernorm.weight": "model-00005-of-00028.safetensors",
+    "gpt_neox.layers.6.mlp.dense_4h_to_h.bias": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.6.mlp.dense_4h_to_h.weight": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.6.mlp.dense_h_to_4h.bias": "model-00005-of-00028.safetensors",
+    "gpt_neox.layers.6.mlp.dense_h_to_4h.weight": "model-00005-of-00028.safetensors",
+    "gpt_neox.layers.6.post_attention_layernorm.bias": "model-00005-of-00028.safetensors",
+    "gpt_neox.layers.6.post_attention_layernorm.weight": "model-00005-of-00028.safetensors",
+    "gpt_neox.layers.7.attention.dense.bias": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.7.attention.dense.weight": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.7.attention.query_key_value.bias": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.7.attention.query_key_value.weight": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.7.attention.rotary_emb.inv_freq": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.7.input_layernorm.bias": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.7.input_layernorm.weight": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.7.mlp.dense_4h_to_h.bias": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.7.mlp.dense_4h_to_h.weight": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.7.mlp.dense_h_to_4h.bias": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.7.mlp.dense_h_to_4h.weight": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.7.post_attention_layernorm.bias": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.7.post_attention_layernorm.weight": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.8.attention.dense.bias": "model-00007-of-00028.safetensors",
+    "gpt_neox.layers.8.attention.dense.weight": "model-00007-of-00028.safetensors",
+    "gpt_neox.layers.8.attention.query_key_value.bias": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.8.attention.query_key_value.weight": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.8.attention.rotary_emb.inv_freq": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.8.input_layernorm.bias": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.8.input_layernorm.weight": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.8.mlp.dense_4h_to_h.bias": "model-00007-of-00028.safetensors",
+    "gpt_neox.layers.8.mlp.dense_4h_to_h.weight": "model-00007-of-00028.safetensors",
+    "gpt_neox.layers.8.mlp.dense_h_to_4h.bias": "model-00007-of-00028.safetensors",
+    "gpt_neox.layers.8.mlp.dense_h_to_4h.weight": "model-00007-of-00028.safetensors",
+    "gpt_neox.layers.8.post_attention_layernorm.bias": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.8.post_attention_layernorm.weight": "model-00006-of-00028.safetensors",
+    "gpt_neox.layers.9.attention.dense.bias": "model-00007-of-00028.safetensors",
+    "gpt_neox.layers.9.attention.dense.weight": "model-00007-of-00028.safetensors",
+    "gpt_neox.layers.9.attention.query_key_value.bias": "model-00007-of-00028.safetensors",
+    "gpt_neox.layers.9.attention.query_key_value.weight": "model-00007-of-00028.safetensors",
+    "gpt_neox.layers.9.attention.rotary_emb.inv_freq": "model-00007-of-00028.safetensors",
+    "gpt_neox.layers.9.input_layernorm.bias": "model-00007-of-00028.safetensors",
+    "gpt_neox.layers.9.input_layernorm.weight": "model-00007-of-00028.safetensors",
+    "gpt_neox.layers.9.mlp.dense_4h_to_h.bias": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.9.mlp.dense_4h_to_h.weight": "model-00008-of-00028.safetensors",
+    "gpt_neox.layers.9.mlp.dense_h_to_4h.bias": "model-00007-of-00028.safetensors",
+    "gpt_neox.layers.9.mlp.dense_h_to_4h.weight": "model-00007-of-00028.safetensors",
+    "gpt_neox.layers.9.post_attention_layernorm.bias": "model-00007-of-00028.safetensors",
+    "gpt_neox.layers.9.post_attention_layernorm.weight": "model-00007-of-00028.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,11 @@
+{
+  "additional_special_tokens": [
+    "<|endoftext|>",
+    "<|sep|>",
+    "<|acc|>",
+    "<|tel|>",
+    "<|rrn|>"
+  ],
+  "eos_token": "<|endoftext|>",
+  "pad_token": "<|endoftext|>"
+}

tokenizer_config.json

@@ -0,0 +1,72 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<|unused0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<|unused1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<|sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "30000": {
+      "content": "<|acc|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "30001": {
+      "content": "<|tel|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "30002": {
+      "content": "<|rrn|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|endoftext|>",
+    "<|sep|>",
+    "<|acc|>",
+    "<|tel|>",
+    "<|rrn|>"
+  ],
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|endoftext|>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}