init main model

adapter_config.json

@@ -1,22 +0,0 @@
-{
-  "base_model_name_or_path": "tiiuae/falcon-40b",
-  "bias": "none",
-  "fan_in_fan_out": false,
-  "inference_mode": true,
-  "init_lora_weights": true,
-  "layers_pattern": null,
-  "layers_to_transform": null,
-  "lora_alpha": 32,
-  "lora_dropout": 0.1,
-  "modules_to_save": null,
-  "peft_type": "LORA",
-  "r": 4,
-  "revision": null,
-  "target_modules": [
-    "query_key_value",
-    "dense",
-    "dense_h_to_4h",
-    "dense_4h_to_h"
-  ],
-  "task_type": "CAUSAL_LM"
-}

config.json

@@ -0,0 +1,33 @@
+{
+  "_name_or_path": "tiiuae/falcon-40b",
+  "alibi": false,
+  "apply_residual_connection_post_layernorm": false,
+  "architectures": [
+    "RWForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "tiiuae/falcon-40b--configuration_RW.RWConfig",
+    "AutoModel": "tiiuae/falcon-40b--modelling_RW.RWModel",
+    "AutoModelForCausalLM": "tiiuae/falcon-40b--modelling_RW.RWForCausalLM",
+    "AutoModelForQuestionAnswering": "tiiuae/falcon-40b--modelling_RW.RWForQuestionAnswering",
+    "AutoModelForSequenceClassification": "tiiuae/falcon-40b--modelling_RW.RWForSequenceClassification",
+    "AutoModelForTokenClassification": "tiiuae/falcon-40b--modelling_RW.RWForTokenClassification"
+  },
+  "bias": false,
+  "bos_token_id": 11,
+  "eos_token_id": 11,
+  "hidden_dropout": 0.0,
+  "hidden_size": 8192,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "RefinedWeb",
+  "n_head": 128,
+  "n_head_kv": 8,
+  "n_layer": 60,
+  "parallel_attn": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.30.1",
+  "use_cache": true,
+  "vocab_size": 65024
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.30.1"
+}

adapter_model.bin → model-00001-of-00009.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:987670579cbb7749ef24514721dab6a9e950722b238123ab455708dd05931a72
-size 55621053
+oid sha256:ece4914ad5646f2ab66cbe851e39be96582042116a40442eaafa0fac4d7eb24f
+size 9504757808

model-00002-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e746b0b0bc97b3a0ae6c7f9300112c14b88b2c2dc0614469408afa6b76d3d5b8
+size 9513146576

model-00003-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d80c14bf59de2a80e25c56d0ed29a9935b6b41aa09c05e18631da87616eb58f
+size 9513146640

model-00004-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56cf81a3c48bf932d677297f54b248baae613acf4083cd79b57d5655b3c0c704
+size 9513146640

model-00005-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d23765968b4f239e64665bf862d4fc0e6f6b6b82de20e577e3a86156597db807
+size 9513146640

model-00006-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e22d2102570c38f317e42adb8291f92899b8fe59c33d624a99c42cc6adab4f4
+size 9513146640

model-00007-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c018e1ee0a178ce10ec7f7be023ce05b12a0441005620c466990efc49a02da1f
+size 9513146640

model-00008-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb902ff91071c021ac0ea5335cd00b800babad01fbbf5b34114633138995fa3f
+size 9513146640

model-00009-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52504b0597bc95565f8994cc8a12cb70b207b58d1ebc2cbf836cdb53149a2d14
+size 6509858832

model.safetensors.index.json

@@ -0,0 +1,490 @@
+{
+  "metadata": {
+    "total_size": 82606587904
+  },
+  "weight_map": {
+    "transformer.h.0.ln_attn.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.0.ln_attn.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.0.ln_mlp.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.0.ln_mlp.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.0.mlp.dense_4h_to_h.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.0.mlp.dense_h_to_4h.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.0.self_attention.dense.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.0.self_attention.query_key_value.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.1.ln_attn.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.1.ln_attn.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.1.ln_mlp.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.1.ln_mlp.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.1.mlp.dense_4h_to_h.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.1.mlp.dense_h_to_4h.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.1.self_attention.dense.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.1.self_attention.query_key_value.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.10.ln_attn.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.10.ln_attn.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.10.ln_mlp.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.10.ln_mlp.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.10.mlp.dense_4h_to_h.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.10.mlp.dense_h_to_4h.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.10.self_attention.dense.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.10.self_attention.query_key_value.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.11.ln_attn.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.11.ln_attn.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.11.ln_mlp.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.11.ln_mlp.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.11.mlp.dense_4h_to_h.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.11.mlp.dense_h_to_4h.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.11.self_attention.dense.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.11.self_attention.query_key_value.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.12.ln_attn.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.12.ln_attn.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.12.ln_mlp.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.12.ln_mlp.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.12.mlp.dense_4h_to_h.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.12.mlp.dense_h_to_4h.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.12.self_attention.dense.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.12.self_attention.query_key_value.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.13.ln_attn.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.13.ln_attn.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.13.ln_mlp.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.13.ln_mlp.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.13.mlp.dense_4h_to_h.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.13.mlp.dense_h_to_4h.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.13.self_attention.dense.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.13.self_attention.query_key_value.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.14.ln_attn.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.14.ln_attn.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.14.ln_mlp.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.14.ln_mlp.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.14.mlp.dense_4h_to_h.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.14.mlp.dense_h_to_4h.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.14.self_attention.dense.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.14.self_attention.query_key_value.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.15.ln_attn.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.15.ln_attn.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.15.ln_mlp.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.15.ln_mlp.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.15.mlp.dense_4h_to_h.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.15.mlp.dense_h_to_4h.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.15.self_attention.dense.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.15.self_attention.query_key_value.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.16.ln_attn.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.16.ln_attn.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.16.ln_mlp.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.16.ln_mlp.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.16.mlp.dense_4h_to_h.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.16.mlp.dense_h_to_4h.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.16.self_attention.dense.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.16.self_attention.query_key_value.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.17.ln_attn.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.17.ln_attn.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.17.ln_mlp.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.17.ln_mlp.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.17.mlp.dense_4h_to_h.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.17.mlp.dense_h_to_4h.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.17.self_attention.dense.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.17.self_attention.query_key_value.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.18.ln_attn.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.18.ln_attn.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.18.ln_mlp.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.18.ln_mlp.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.18.mlp.dense_4h_to_h.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.18.mlp.dense_h_to_4h.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.18.self_attention.dense.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.18.self_attention.query_key_value.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.19.ln_attn.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.19.ln_attn.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.19.ln_mlp.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.19.ln_mlp.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.19.mlp.dense_4h_to_h.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.19.mlp.dense_h_to_4h.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.19.self_attention.dense.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.19.self_attention.query_key_value.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.2.ln_attn.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.2.ln_attn.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.2.ln_mlp.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.2.ln_mlp.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.2.mlp.dense_4h_to_h.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.2.mlp.dense_h_to_4h.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.2.self_attention.dense.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.2.self_attention.query_key_value.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.20.ln_attn.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.20.ln_attn.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.20.ln_mlp.bias": "model-00003-of-00009.safetensors",
+    "transformer.h.20.ln_mlp.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.20.mlp.dense_4h_to_h.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.20.mlp.dense_h_to_4h.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.20.self_attention.dense.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.20.self_attention.query_key_value.weight": "model-00003-of-00009.safetensors",
+    "transformer.h.21.ln_attn.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.21.ln_attn.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.21.ln_mlp.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.21.ln_mlp.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.21.mlp.dense_4h_to_h.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.21.mlp.dense_h_to_4h.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.21.self_attention.dense.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.21.self_attention.query_key_value.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.22.ln_attn.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.22.ln_attn.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.22.ln_mlp.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.22.ln_mlp.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.22.mlp.dense_4h_to_h.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.22.mlp.dense_h_to_4h.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.22.self_attention.dense.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.22.self_attention.query_key_value.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.23.ln_attn.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.23.ln_attn.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.23.ln_mlp.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.23.ln_mlp.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.23.mlp.dense_4h_to_h.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.23.mlp.dense_h_to_4h.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.23.self_attention.dense.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.23.self_attention.query_key_value.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.24.ln_attn.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.24.ln_attn.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.24.ln_mlp.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.24.ln_mlp.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.24.mlp.dense_4h_to_h.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.24.mlp.dense_h_to_4h.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.24.self_attention.dense.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.24.self_attention.query_key_value.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.25.ln_attn.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.25.ln_attn.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.25.ln_mlp.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.25.ln_mlp.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.25.mlp.dense_4h_to_h.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.25.mlp.dense_h_to_4h.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.25.self_attention.dense.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.25.self_attention.query_key_value.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.26.ln_attn.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.26.ln_attn.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.26.ln_mlp.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.26.ln_mlp.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.26.mlp.dense_4h_to_h.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.26.mlp.dense_h_to_4h.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.26.self_attention.dense.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.26.self_attention.query_key_value.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.27.ln_attn.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.27.ln_attn.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.27.ln_mlp.bias": "model-00004-of-00009.safetensors",
+    "transformer.h.27.ln_mlp.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.27.mlp.dense_4h_to_h.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.27.mlp.dense_h_to_4h.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.27.self_attention.dense.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.27.self_attention.query_key_value.weight": "model-00004-of-00009.safetensors",
+    "transformer.h.28.ln_attn.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.28.ln_attn.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.28.ln_mlp.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.28.ln_mlp.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.28.mlp.dense_4h_to_h.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.28.mlp.dense_h_to_4h.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.28.self_attention.dense.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.28.self_attention.query_key_value.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.29.ln_attn.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.29.ln_attn.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.29.ln_mlp.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.29.ln_mlp.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.29.mlp.dense_4h_to_h.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.29.mlp.dense_h_to_4h.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.29.self_attention.dense.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.29.self_attention.query_key_value.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.3.ln_attn.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.3.ln_attn.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.3.ln_mlp.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.3.ln_mlp.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.3.mlp.dense_4h_to_h.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.3.mlp.dense_h_to_4h.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.3.self_attention.dense.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.3.self_attention.query_key_value.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.30.ln_attn.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.30.ln_attn.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.30.ln_mlp.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.30.ln_mlp.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.30.mlp.dense_4h_to_h.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.30.mlp.dense_h_to_4h.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.30.self_attention.dense.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.30.self_attention.query_key_value.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.31.ln_attn.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.31.ln_attn.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.31.ln_mlp.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.31.ln_mlp.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.31.mlp.dense_4h_to_h.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.31.mlp.dense_h_to_4h.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.31.self_attention.dense.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.31.self_attention.query_key_value.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.32.ln_attn.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.32.ln_attn.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.32.ln_mlp.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.32.ln_mlp.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.32.mlp.dense_4h_to_h.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.32.mlp.dense_h_to_4h.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.32.self_attention.dense.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.32.self_attention.query_key_value.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.33.ln_attn.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.33.ln_attn.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.33.ln_mlp.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.33.ln_mlp.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.33.mlp.dense_4h_to_h.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.33.mlp.dense_h_to_4h.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.33.self_attention.dense.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.33.self_attention.query_key_value.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.34.ln_attn.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.34.ln_attn.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.34.ln_mlp.bias": "model-00005-of-00009.safetensors",
+    "transformer.h.34.ln_mlp.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.34.mlp.dense_4h_to_h.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.34.mlp.dense_h_to_4h.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.34.self_attention.dense.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.34.self_attention.query_key_value.weight": "model-00005-of-00009.safetensors",
+    "transformer.h.35.ln_attn.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.35.ln_attn.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.35.ln_mlp.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.35.ln_mlp.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.35.mlp.dense_4h_to_h.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.35.mlp.dense_h_to_4h.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.35.self_attention.dense.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.35.self_attention.query_key_value.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.36.ln_attn.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.36.ln_attn.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.36.ln_mlp.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.36.ln_mlp.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.36.mlp.dense_4h_to_h.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.36.mlp.dense_h_to_4h.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.36.self_attention.dense.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.36.self_attention.query_key_value.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.37.ln_attn.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.37.ln_attn.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.37.ln_mlp.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.37.ln_mlp.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.37.mlp.dense_4h_to_h.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.37.mlp.dense_h_to_4h.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.37.self_attention.dense.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.37.self_attention.query_key_value.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.38.ln_attn.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.38.ln_attn.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.38.ln_mlp.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.38.ln_mlp.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.38.mlp.dense_4h_to_h.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.38.mlp.dense_h_to_4h.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.38.self_attention.dense.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.38.self_attention.query_key_value.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.39.ln_attn.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.39.ln_attn.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.39.ln_mlp.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.39.ln_mlp.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.39.mlp.dense_4h_to_h.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.39.mlp.dense_h_to_4h.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.39.self_attention.dense.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.39.self_attention.query_key_value.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.4.ln_attn.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.4.ln_attn.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.4.ln_mlp.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.4.ln_mlp.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.4.mlp.dense_4h_to_h.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.4.mlp.dense_h_to_4h.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.4.self_attention.dense.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.4.self_attention.query_key_value.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.40.ln_attn.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.40.ln_attn.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.40.ln_mlp.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.40.ln_mlp.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.40.mlp.dense_4h_to_h.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.40.mlp.dense_h_to_4h.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.40.self_attention.dense.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.40.self_attention.query_key_value.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.41.ln_attn.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.41.ln_attn.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.41.ln_mlp.bias": "model-00006-of-00009.safetensors",
+    "transformer.h.41.ln_mlp.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.41.mlp.dense_4h_to_h.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.41.mlp.dense_h_to_4h.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.41.self_attention.dense.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.41.self_attention.query_key_value.weight": "model-00006-of-00009.safetensors",
+    "transformer.h.42.ln_attn.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.42.ln_attn.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.42.ln_mlp.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.42.ln_mlp.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.42.mlp.dense_4h_to_h.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.42.mlp.dense_h_to_4h.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.42.self_attention.dense.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.42.self_attention.query_key_value.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.43.ln_attn.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.43.ln_attn.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.43.ln_mlp.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.43.ln_mlp.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.43.mlp.dense_4h_to_h.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.43.mlp.dense_h_to_4h.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.43.self_attention.dense.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.43.self_attention.query_key_value.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.44.ln_attn.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.44.ln_attn.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.44.ln_mlp.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.44.ln_mlp.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.44.mlp.dense_4h_to_h.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.44.mlp.dense_h_to_4h.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.44.self_attention.dense.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.44.self_attention.query_key_value.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.45.ln_attn.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.45.ln_attn.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.45.ln_mlp.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.45.ln_mlp.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.45.mlp.dense_4h_to_h.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.45.mlp.dense_h_to_4h.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.45.self_attention.dense.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.45.self_attention.query_key_value.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.46.ln_attn.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.46.ln_attn.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.46.ln_mlp.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.46.ln_mlp.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.46.mlp.dense_4h_to_h.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.46.mlp.dense_h_to_4h.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.46.self_attention.dense.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.46.self_attention.query_key_value.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.47.ln_attn.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.47.ln_attn.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.47.ln_mlp.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.47.ln_mlp.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.47.mlp.dense_4h_to_h.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.47.mlp.dense_h_to_4h.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.47.self_attention.dense.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.47.self_attention.query_key_value.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.48.ln_attn.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.48.ln_attn.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.48.ln_mlp.bias": "model-00007-of-00009.safetensors",
+    "transformer.h.48.ln_mlp.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.48.mlp.dense_4h_to_h.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.48.mlp.dense_h_to_4h.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.48.self_attention.dense.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.48.self_attention.query_key_value.weight": "model-00007-of-00009.safetensors",
+    "transformer.h.49.ln_attn.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.49.ln_attn.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.49.ln_mlp.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.49.ln_mlp.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.49.mlp.dense_4h_to_h.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.49.mlp.dense_h_to_4h.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.49.self_attention.dense.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.49.self_attention.query_key_value.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.5.ln_attn.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.5.ln_attn.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.5.ln_mlp.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.5.ln_mlp.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.5.mlp.dense_4h_to_h.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.5.mlp.dense_h_to_4h.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.5.self_attention.dense.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.5.self_attention.query_key_value.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.50.ln_attn.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.50.ln_attn.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.50.ln_mlp.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.50.ln_mlp.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.50.mlp.dense_4h_to_h.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.50.mlp.dense_h_to_4h.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.50.self_attention.dense.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.50.self_attention.query_key_value.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.51.ln_attn.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.51.ln_attn.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.51.ln_mlp.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.51.ln_mlp.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.51.mlp.dense_4h_to_h.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.51.mlp.dense_h_to_4h.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.51.self_attention.dense.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.51.self_attention.query_key_value.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.52.ln_attn.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.52.ln_attn.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.52.ln_mlp.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.52.ln_mlp.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.52.mlp.dense_4h_to_h.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.52.mlp.dense_h_to_4h.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.52.self_attention.dense.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.52.self_attention.query_key_value.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.53.ln_attn.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.53.ln_attn.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.53.ln_mlp.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.53.ln_mlp.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.53.mlp.dense_4h_to_h.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.53.mlp.dense_h_to_4h.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.53.self_attention.dense.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.53.self_attention.query_key_value.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.54.ln_attn.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.54.ln_attn.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.54.ln_mlp.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.54.ln_mlp.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.54.mlp.dense_4h_to_h.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.54.mlp.dense_h_to_4h.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.54.self_attention.dense.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.54.self_attention.query_key_value.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.55.ln_attn.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.55.ln_attn.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.55.ln_mlp.bias": "model-00008-of-00009.safetensors",
+    "transformer.h.55.ln_mlp.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.55.mlp.dense_4h_to_h.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.55.mlp.dense_h_to_4h.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.55.self_attention.dense.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.55.self_attention.query_key_value.weight": "model-00008-of-00009.safetensors",
+    "transformer.h.56.ln_attn.bias": "model-00009-of-00009.safetensors",
+    "transformer.h.56.ln_attn.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.56.ln_mlp.bias": "model-00009-of-00009.safetensors",
+    "transformer.h.56.ln_mlp.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.56.mlp.dense_4h_to_h.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.56.mlp.dense_h_to_4h.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.56.self_attention.dense.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.56.self_attention.query_key_value.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.57.ln_attn.bias": "model-00009-of-00009.safetensors",
+    "transformer.h.57.ln_attn.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.57.ln_mlp.bias": "model-00009-of-00009.safetensors",
+    "transformer.h.57.ln_mlp.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.57.mlp.dense_4h_to_h.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.57.mlp.dense_h_to_4h.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.57.self_attention.dense.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.57.self_attention.query_key_value.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.58.ln_attn.bias": "model-00009-of-00009.safetensors",
+    "transformer.h.58.ln_attn.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.58.ln_mlp.bias": "model-00009-of-00009.safetensors",
+    "transformer.h.58.ln_mlp.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.58.mlp.dense_4h_to_h.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.58.mlp.dense_h_to_4h.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.58.self_attention.dense.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.58.self_attention.query_key_value.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.59.ln_attn.bias": "model-00009-of-00009.safetensors",
+    "transformer.h.59.ln_attn.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.59.ln_mlp.bias": "model-00009-of-00009.safetensors",
+    "transformer.h.59.ln_mlp.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.59.mlp.dense_4h_to_h.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.59.mlp.dense_h_to_4h.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.59.self_attention.dense.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.59.self_attention.query_key_value.weight": "model-00009-of-00009.safetensors",
+    "transformer.h.6.ln_attn.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.6.ln_attn.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.6.ln_mlp.bias": "model-00001-of-00009.safetensors",
+    "transformer.h.6.ln_mlp.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.6.mlp.dense_4h_to_h.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.6.mlp.dense_h_to_4h.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.6.self_attention.dense.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.6.self_attention.query_key_value.weight": "model-00001-of-00009.safetensors",
+    "transformer.h.7.ln_attn.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.7.ln_attn.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.7.ln_mlp.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.7.ln_mlp.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.7.mlp.dense_4h_to_h.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.7.mlp.dense_h_to_4h.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.7.self_attention.dense.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.7.self_attention.query_key_value.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.8.ln_attn.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.8.ln_attn.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.8.ln_mlp.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.8.ln_mlp.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.8.mlp.dense_4h_to_h.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.8.mlp.dense_h_to_4h.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.8.self_attention.dense.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.8.self_attention.query_key_value.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.9.ln_attn.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.9.ln_attn.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.9.ln_mlp.bias": "model-00002-of-00009.safetensors",
+    "transformer.h.9.ln_mlp.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.9.mlp.dense_4h_to_h.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.9.mlp.dense_h_to_4h.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.9.self_attention.dense.weight": "model-00002-of-00009.safetensors",
+    "transformer.h.9.self_attention.query_key_value.weight": "model-00002-of-00009.safetensors",
+    "transformer.ln_f.bias": "model-00009-of-00009.safetensors",
+    "transformer.ln_f.weight": "model-00009-of-00009.safetensors",
+    "transformer.word_embeddings.weight": "model-00001-of-00009.safetensors"
+  }
+}

special_tokens_map.json

@@ -12,6 +12,5 @@
     ">>SUFFIX<<",
     ">>MIDDLE<<"
   ],
-  "eos_token": "<|endoftext|>",
-  "pad_token": "<|endoftext|>"
+  "eos_token": "<|endoftext|>"
 }