Update Model

config.json

@@ -1,5 +1,5 @@
 {
-    "_name_or_path": "barchetta/barchetta-naso-22191733",
+    "_name_or_path": "workspace/trained-models/quickstep3621/nigmqxw",
     "architectures": [
         "LlamaForCausalLM"
     ],
@@ -18,6 +18,7 @@
     "num_attention_heads": 32,
     "num_hidden_layers": 32,
     "num_key_value_heads": 8,
+    "pad_token_id": 128004,
     "pretraining_tp": 1,
     "rms_norm_eps": 1.0e-5,
     "rope_scaling": {
@@ -31,6 +32,7 @@
     "tie_word_embeddings": false,
     "torch_dtype": "bfloat16",
     "transformers_version": "4.48.1",
+    "unsloth_version": "2025.1.6",
     "use_cache": false,
     "vocab_size": 128258
-}
+}

generation_config.json

@@ -2,6 +2,8 @@
   "_from_model_config": true,
   "bos_token_id": 128000,
   "eos_token_id": 128001,
+  "max_length": 131072,
+  "pad_token_id": 128004,
   "transformers_version": "4.48.1",
   "use_cache": false
 }

model-00001-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c1e6b51d89426425533cf7c961d6b7298ff9f43416631cda80ff85a115333479
-size 4976715104
+oid sha256:e5fe143515e69d9be77969a732c7b9a835583edca0c13359d0ecc158b436160d
+size 4976715096

model-00002-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:20f770b1e4c06439a3a4f8c61ca2733ca686e304695dc7c0525a24996832d6bf
-size 4999802768
+oid sha256:dc7eb066e73dce3242dfa415da682613419202bfef5e9dcb6ddf087716f9b7c9
+size 4999802760

model-00003-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:17d05434e57b5388741584702dce70027389091c6be5c4e89d469ea14f61faac
-size 4915916232
+oid sha256:508baa46b1aa276aaacf8e83a41fd140fa0f2431bfeb36b9e7221fa3c25acde3
+size 4915916216

model-00004-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4f9d828e32d2bad0b5559cfd086b6eb2ca19bceb8b7e848dcad7f8c55284a6b6
-size 1168155240
+oid sha256:99e3e55db49458431710eab2d81ed874cf96747be0d4e6a5f097ed08fac657af
+size 1168155232

model.safetensors.index.json

@@ -1,6 +1,7 @@
 {
   "metadata": {
-    "total_size": 16060555264
+    "total_size": 16060555264,
+    "path": "workspace/trained-models/quickstep3621/nigmqxw"
   },
   "weight_map": {
     "lm_head.weight": "model-00004-of-00004.safetensors",

special_tokens_map.json

@@ -17,11 +17,5 @@
     "rstrip": false,
     "single_word": false
   },
-  "pad_token": {
-    "content": "<|im_end|>",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
-    "single_word": false
-  }
+  "pad_token": "<|finetune_right_pad_id|>"
 }

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:510d12ec255f4cb0304aa5428d699c354c1a49696b427a2748a7b03bb7bbb575
-size 17210296
+oid sha256:eda94f467b2242e7aefa3816ab89a430a7df755e340b598814c87fb7c92633a4
+size 17210395

tokenizer_config.json

@@ -2079,6 +2079,7 @@
     "attention_mask"
   ],
   "model_max_length": 131072,
-  "pad_token": "<|im_end|>",
+  "pad_token": "<|finetune_right_pad_id|>",
+  "padding_side": "left",
   "tokenizer_class": "PreTrainedTokenizerFast"
 }