Add 8bit quantized model

all_results.json

@@ -1,8 +0,0 @@
-{
-    "epoch": 1.99,
-    "train_loss": 0.7372576573065349,
-    "train_runtime": 28706.0932,
-    "train_samples": 57593,
-    "train_samples_per_second": 4.013,
-    "train_steps_per_second": 0.016
-}

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "EleutherAI/polyglot-ko-12.8b",
+  "_name_or_path": "beomi/KoAlpaca-Polyglot-12.8B",
   "architectures": [
     "GPTNeoXForCausalLM"
   ],
@@ -15,11 +15,17 @@
   "num_attention_heads": 40,
   "num_hidden_layers": 40,
   "num_steps": "global_step301000",
+  "quantization_config": {
+    "llm_int8_enable_fp32_cpu_offload": false,
+    "llm_int8_skip_modules": null,
+    "llm_int8_threshold": 6.0,
+    "load_in_8bit": true
+  },
   "rotary_emb_base": 10000,
   "rotary_pct": 0.5,
   "tie_word_embeddings": false,
   "torch_dtype": "float16",
-  "transformers_version": "4.28.1",
+  "transformers_version": "4.29.0.dev0",
   "use_cache": true,
   "use_parallel_residual": true,
   "vocab_size": 30003

generation_config.json

@@ -2,5 +2,5 @@
   "_from_model_config": true,
   "bos_token_id": 0,
   "eos_token_id": 0,
-  "transformers_version": "4.28.1"
+  "transformers_version": "4.29.0.dev0"
 }

pytorch_model.bin → pytorch_model-00001-of-00002.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4b06c217dc641008ad56030c95d59940d9fd021d87a9c1321e6a5bc721bcc6ed
-size 25953607721
+oid sha256:fd8b18b742a5eb3f3d26f7096c210f7864df9e69e53861fdaf10b1942d8debdc
+size 10094176536

training_args.bin → pytorch_model-00002-of-00002.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d16517ebe6740e1b5ae9b137c22d1caae3fb8e1663fe042699314cf61bc64831
-size 4859
+oid sha256:04a3a6330f8ad6716fd5f3ce52f3dfca65ff910cc16bbe39a6cdb4d612c6867f
+size 3283963278

pytorch_model.bin.index.json

@@ -0,0 +1,771 @@
+{
+  "metadata": {
+    "total_size": 13231065680.0
+  },
+  "weight_map": {
+    "embed_out.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.embed_in.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.final_layer_norm.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.final_layer_norm.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.0.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.0.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.0.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.0.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.0.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.0.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.0.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.0.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.0.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.0.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.0.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.0.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.0.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.0.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.0.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.0.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.0.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.0.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.0.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.1.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.1.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.1.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.1.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.1.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.1.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.1.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.1.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.1.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.1.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.1.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.1.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.1.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.1.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.1.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.1.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.1.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.1.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.1.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.10.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.10.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.10.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.10.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.10.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.10.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.10.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.10.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.10.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.10.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.10.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.10.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.10.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.10.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.10.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.10.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.10.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.10.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.10.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.11.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.11.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.11.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.11.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.11.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.11.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.11.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.11.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.11.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.11.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.11.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.11.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.11.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.11.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.11.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.11.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.11.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.11.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.11.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.12.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.12.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.12.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.12.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.12.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.12.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.12.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.12.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.12.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.12.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.12.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.12.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.12.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.12.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.12.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.12.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.12.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.12.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.12.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.13.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.13.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.13.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.13.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.13.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.13.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.13.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.13.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.13.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.13.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.13.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.13.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.13.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.13.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.13.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.13.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.13.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.13.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.13.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.14.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.14.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.14.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.14.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.14.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.14.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.14.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.14.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.14.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.14.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.14.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.14.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.14.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.14.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.14.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.14.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.14.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.14.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.14.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.15.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.15.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.15.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.15.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.15.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.15.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.15.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.15.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.15.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.15.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.15.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.15.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.15.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.15.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.15.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.15.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.15.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.15.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.15.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.16.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.16.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.16.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.16.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.16.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.16.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.16.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.16.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.16.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.16.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.16.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.16.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.16.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.16.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.16.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.16.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.16.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.16.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.16.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.17.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.17.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.17.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.17.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.17.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.17.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.17.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.17.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.17.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.17.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.17.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.17.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.17.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.17.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.17.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.17.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.17.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.17.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.17.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.18.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.18.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.18.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.18.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.18.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.18.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.18.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.18.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.18.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.18.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.18.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.18.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.18.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.18.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.18.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.18.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.18.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.18.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.18.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.19.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.19.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.19.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.19.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.19.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.19.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.19.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.19.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.19.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.19.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.19.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.19.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.19.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.19.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.19.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.19.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.19.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.19.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.19.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.2.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.2.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.2.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.2.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.2.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.2.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.2.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.2.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.2.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.2.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.2.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.2.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.2.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.2.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.2.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.2.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.2.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.2.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.2.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.20.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.20.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.20.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.20.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.20.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.20.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.20.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.20.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.20.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.20.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.20.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.20.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.20.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.20.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.20.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.20.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.20.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.20.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.20.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.21.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.21.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.21.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.21.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.21.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.21.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.21.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.21.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.21.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.21.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.21.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.21.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.21.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.21.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.21.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.21.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.21.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.21.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.21.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.22.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.22.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.22.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.22.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.22.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.22.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.22.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.22.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.22.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.22.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.22.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.22.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.22.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.22.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.22.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.22.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.22.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.22.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.22.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.23.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.23.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.23.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.23.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.23.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.23.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.23.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.23.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.23.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.23.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.23.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.23.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.23.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.23.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.23.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.23.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.23.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.23.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.23.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.24.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.24.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.24.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.24.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.24.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.24.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.24.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.24.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.24.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.24.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.24.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.24.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.24.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.24.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.24.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.24.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.24.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.24.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.24.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.25.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.25.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.25.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.25.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.25.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.25.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.25.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.25.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.25.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.25.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.25.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.25.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.25.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.25.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.25.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.25.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.25.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.25.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.25.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.26.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.26.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.26.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.26.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.26.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.26.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.26.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.26.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.26.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.26.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.26.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.26.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.26.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.26.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.26.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.26.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.26.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.26.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.26.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.27.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.27.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.27.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.27.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.27.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.27.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.27.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.27.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.27.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.27.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.27.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.27.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.27.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.27.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.27.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.27.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.27.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.27.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.27.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.28.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.28.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.28.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.28.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.28.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.28.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.28.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.28.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.28.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.28.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.28.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.28.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.28.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.28.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.28.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.28.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.28.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.28.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.28.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.29.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.29.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.29.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.29.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.29.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.29.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.29.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.29.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.29.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.29.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.29.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.29.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.29.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.29.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.29.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.29.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.29.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.29.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.29.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.3.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.3.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.3.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.3.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.3.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.3.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.3.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.3.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.3.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.3.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.3.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.3.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.3.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.3.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.3.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.3.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.3.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.3.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.3.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.30.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.30.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.30.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.30.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.30.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.30.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.30.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.30.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.30.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.30.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.30.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.30.mlp.dense_4h_to_h.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.30.mlp.dense_4h_to_h.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.30.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.30.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.30.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.30.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.30.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.30.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.31.attention.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.31.attention.dense.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.31.attention.dense.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.31.attention.dense.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.31.attention.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.31.attention.query_key_value.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.31.attention.query_key_value.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.31.attention.query_key_value.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.31.attention.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.31.input_layernorm.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.31.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.31.mlp.dense_4h_to_h.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.31.mlp.dense_4h_to_h.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.31.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.31.mlp.dense_h_to_4h.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.31.mlp.dense_h_to_4h.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.31.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.31.post_attention_layernorm.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.31.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.32.attention.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.32.attention.dense.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.32.attention.dense.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.32.attention.dense.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.32.attention.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.32.attention.query_key_value.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.32.attention.query_key_value.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.32.attention.query_key_value.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.32.attention.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.32.input_layernorm.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.32.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.32.mlp.dense_4h_to_h.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.32.mlp.dense_4h_to_h.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.32.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.32.mlp.dense_h_to_4h.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.32.mlp.dense_h_to_4h.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.32.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.32.post_attention_layernorm.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.32.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.33.attention.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.33.attention.dense.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.33.attention.dense.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.33.attention.dense.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.33.attention.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.33.attention.query_key_value.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.33.attention.query_key_value.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.33.attention.query_key_value.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.33.attention.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.33.input_layernorm.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.33.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.33.mlp.dense_4h_to_h.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.33.mlp.dense_4h_to_h.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.33.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.33.mlp.dense_h_to_4h.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.33.mlp.dense_h_to_4h.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.33.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.33.post_attention_layernorm.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.33.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.34.attention.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.34.attention.dense.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.34.attention.dense.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.34.attention.dense.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.34.attention.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.34.attention.query_key_value.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.34.attention.query_key_value.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.34.attention.query_key_value.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.34.attention.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.34.input_layernorm.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.34.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.34.mlp.dense_4h_to_h.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.34.mlp.dense_4h_to_h.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.34.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.34.mlp.dense_h_to_4h.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.34.mlp.dense_h_to_4h.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.34.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.34.post_attention_layernorm.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.34.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.35.attention.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.35.attention.dense.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.35.attention.dense.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.35.attention.dense.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.35.attention.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.35.attention.query_key_value.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.35.attention.query_key_value.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.35.attention.query_key_value.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.35.attention.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.35.input_layernorm.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.35.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.35.mlp.dense_4h_to_h.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.35.mlp.dense_4h_to_h.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.35.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.35.mlp.dense_h_to_4h.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.35.mlp.dense_h_to_4h.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.35.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.35.post_attention_layernorm.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.35.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.36.attention.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.36.attention.dense.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.36.attention.dense.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.36.attention.dense.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.36.attention.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.36.attention.query_key_value.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.36.attention.query_key_value.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.36.attention.query_key_value.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.36.attention.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.36.input_layernorm.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.36.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.36.mlp.dense_4h_to_h.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.36.mlp.dense_4h_to_h.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.36.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.36.mlp.dense_h_to_4h.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.36.mlp.dense_h_to_4h.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.36.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.36.post_attention_layernorm.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.36.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.37.attention.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.37.attention.dense.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.37.attention.dense.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.37.attention.dense.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.37.attention.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.37.attention.query_key_value.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.37.attention.query_key_value.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.37.attention.query_key_value.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.37.attention.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.37.input_layernorm.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.37.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.37.mlp.dense_4h_to_h.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.37.mlp.dense_4h_to_h.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.37.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.37.mlp.dense_h_to_4h.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.37.mlp.dense_h_to_4h.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.37.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.37.post_attention_layernorm.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.37.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.38.attention.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.38.attention.dense.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.38.attention.dense.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.38.attention.dense.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.38.attention.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.38.attention.query_key_value.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.38.attention.query_key_value.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.38.attention.query_key_value.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.38.attention.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.38.input_layernorm.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.38.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.38.mlp.dense_4h_to_h.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.38.mlp.dense_4h_to_h.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.38.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.38.mlp.dense_h_to_4h.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.38.mlp.dense_h_to_4h.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.38.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.38.post_attention_layernorm.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.38.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.39.attention.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.39.attention.dense.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.39.attention.dense.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.39.attention.dense.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.39.attention.masked_bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.39.attention.query_key_value.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.39.attention.query_key_value.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.39.attention.query_key_value.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.39.attention.rotary_emb.inv_freq": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.39.input_layernorm.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.39.input_layernorm.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.39.mlp.dense_4h_to_h.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.39.mlp.dense_4h_to_h.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.39.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.39.mlp.dense_h_to_4h.SCB": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.39.mlp.dense_h_to_4h.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.39.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.39.post_attention_layernorm.bias": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.39.post_attention_layernorm.weight": "pytorch_model-00002-of-00002.bin",
+    "gpt_neox.layers.4.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.4.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.4.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.4.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.4.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.4.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.4.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.4.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.4.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.4.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.4.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.4.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.4.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.4.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.4.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.4.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.4.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.4.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.4.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.5.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.5.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.5.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.5.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.5.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.5.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.5.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.5.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.5.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.5.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.5.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.5.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.5.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.5.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.5.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.5.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.5.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.5.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.5.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.6.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.6.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.6.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.6.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.6.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.6.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.6.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.6.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.6.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.6.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.6.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.6.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.6.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.6.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.6.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.6.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.6.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.6.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.6.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.7.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.7.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.7.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.7.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.7.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.7.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.7.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.7.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.7.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.7.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.7.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.7.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.7.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.7.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.7.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.7.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.7.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.7.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.7.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.8.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.8.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.8.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.8.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.8.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.8.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.8.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.8.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.8.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.8.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.8.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.8.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.8.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.8.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.8.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.8.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.8.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.8.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.8.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.9.attention.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.9.attention.dense.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.9.attention.dense.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.9.attention.dense.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.9.attention.masked_bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.9.attention.query_key_value.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.9.attention.query_key_value.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.9.attention.query_key_value.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.9.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.9.input_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.9.input_layernorm.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.9.mlp.dense_4h_to_h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.9.mlp.dense_4h_to_h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.9.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.9.mlp.dense_h_to_4h.SCB": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.9.mlp.dense_h_to_4h.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.9.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.9.post_attention_layernorm.bias": "pytorch_model-00001-of-00002.bin",
+    "gpt_neox.layers.9.post_attention_layernorm.weight": "pytorch_model-00001-of-00002.bin"
+  }
+}

train_results.json

@@ -1,8 +0,0 @@
-{
-    "epoch": 1.99,
-    "train_loss": 0.7372576573065349,
-    "train_runtime": 28706.0932,
-    "train_samples": 57593,
-    "train_samples_per_second": 4.013,
-    "train_steps_per_second": 0.016
-}

trainer_state.json

@@ -1,31 +0,0 @@
-{
-  "best_metric": null,
-  "best_model_checkpoint": null,
-  "epoch": 1.991249392318911,
-  "global_step": 448,
-  "is_hyper_param_search": false,
-  "is_local_process_zero": true,
-  "is_world_process_zero": true,
-  "log_history": [
-    {
-      "epoch": 0.0,
-      "learning_rate": 0,
-      "loss": 2.4259,
-      "step": 1
-    },
-    {
-      "epoch": 1.99,
-      "step": 448,
-      "total_flos": 1883191577149440.0,
-      "train_loss": 0.7372576573065349,
-      "train_runtime": 28706.0932,
-      "train_samples_per_second": 4.013,
-      "train_steps_per_second": 0.016
-    }
-  ],
-  "max_steps": 448,
-  "num_train_epochs": 2,
-  "total_flos": 1883191577149440.0,
-  "trial_name": null,
-  "trial_params": null
-}