Upload LlamaForCausalLM

README.md

@@ -1,10 +1,10 @@
 ---
-library_name: transformers
+language:
+- en
 license: apache-2.0
+library_name: transformers
 datasets:
 - Trelis/orpo-dpo-mix-40k-SHORT
-language:
-- en
 pipeline_tag: text-classification
 ---
 

config.json

@@ -1,12 +1,12 @@
 {
-  "_name_or_path": "meta-llama/Meta-Llama-3-8B",
+  "_name_or_path": "/kaggle/input/llama-3/transformers/8b-hf/1",
   "architectures": [
     "LlamaForCausalLM"
   ],
   "attention_bias": false,
   "attention_dropout": 0.0,
-  "bos_token_id": 128256,
-  "eos_token_id": 128257,
+  "bos_token_id": 128000,
+  "eos_token_id": 128001,
   "hidden_act": "silu",
   "hidden_size": 4096,
   "initializer_range": 0.02,
@@ -17,7 +17,6 @@
   "num_attention_heads": 32,
   "num_hidden_layers": 32,
   "num_key_value_heads": 8,
-  "pad_token_id": 128257,
   "pretraining_tp": 1,
   "rms_norm_eps": 1e-05,
   "rope_scaling": null,
@@ -26,5 +25,5 @@
   "torch_dtype": "float16",
   "transformers_version": "4.41.0",
   "use_cache": true,
-  "vocab_size": 128258
+  "vocab_size": 128256
 }

generation_config.json

@@ -1,10 +1,6 @@
 {
-  "bos_token_id": 128256,
-  "do_sample": true,
-  "eos_token_id": 128257,
-  "max_length": 4096,
-  "pad_token_id": 128257,
-  "temperature": 0.6,
-  "top_p": 0.9,
+  "_from_model_config": true,
+  "bos_token_id": 128000,
+  "eos_token_id": 128001,
   "transformers_version": "4.41.0"
 }

model-00001-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90b105009f6cd276891d0da58573f932960a0ce3b155d9ac5f3d857017ec6ec8
-size 4976714976
+oid sha256:b6948f1132949c27c365b689f126fef076fe4182f553b8c0399639e0135241f4
+size 4976698592

model-00002-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:614751c97e105793b5b9d5cc5a15f23c2c4b00a8cc7b0656ef60b47893a842c8
+oid sha256:414dede41da3a92b825a85b6ee22119d538c20305258f5c6e724626793293179
 size 4999802616

model-00003-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9dd469a74e79763a438c2597f6c53a9b5c80d011bd58e20b636d914677ce1533
+oid sha256:a15f0b0b0857d4017400881f4241d51835ac6c0d622db305c026954141d11faa
 size 4915916080

model-00004-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ac146a09043bcb032f655a3cacb56793eec649fd64ff5e219aa906968e7f141b
-size 1168155192
+oid sha256:350ac0df99121df394b7795107dbe8e4e1beb0ae066b6f8a89e9c504f6553dab
+size 1168138808

model.safetensors.index.json

@@ -1,6 +1,6 @@
 {
   "metadata": {
-    "total_size": 16060555264
+    "total_size": 16060522496
   },
   "weight_map": {
     "lm_head.weight": "model-00004-of-00004.safetensors",