Upload folder using huggingface_hub (#4)

- c1c02e6108986e1cdc215ff9e98e7913aa44170688dc4d9f1fa20736cad647c6 (709b9b2b3681ace42d10bbfdbb36bb0f43408c0b)

config.json

@@ -1,5 +1,5 @@
 {
-    "_name_or_path": "/covalent/.cache/models/tmp8iq7gzkaj8hi32q5",
+    "_name_or_path": "/covalent/.cache/models/tmpqk4aj9ccouxpo9kw",
     "architectures": [
         "LlamaForCausalLM"
     ],
@@ -9,15 +9,15 @@
     "eos_token_id": 2,
     "head_dim": 16,
     "hidden_act": "silu",
-    "hidden_size": 864,
+    "hidden_size": 896,
     "initializer_range": 0.02,
-    "intermediate_size": 3456,
+    "intermediate_size": 3584,
     "max_position_embeddings": 512,
     "mlp_bias": false,
     "model_type": "llama",
-    "num_attention_heads": 54,
-    "num_hidden_layers": 27,
-    "num_key_value_heads": 54,
+    "num_attention_heads": 56,
+    "num_hidden_layers": 28,
+    "num_key_value_heads": 56,
     "pad_token_id": 1,
     "pretraining_tp": 1,
     "quantization_config": {
@@ -44,6 +44,6 @@
     "torch_dtype": "float16",
     "transformers_version": "4.46.2",
     "use_cache": true,
-    "vocab_size": 32000,
+    "vocab_size": 52000,
     "api_key": null
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d1d383e39761d8f1f635756ff90331ed8d135e49149f68577e77693cd08edcc7
-size 434453229
+oid sha256:e3ab9e2f4af434ea41ccbcda53d5e8298a5f58b8e20b1814ca71c13d2d242d65
+size 547505644

smash_config.json

@@ -28,7 +28,7 @@
     "quant_llm-int8_weight_bits": 8,
     "max_batch_size": 1,
     "device": "cuda",
-    "cache_dir": "/covalent/.cache/models/tmp8iq7gzka",
+    "cache_dir": "/covalent/.cache/models/tmpqk4aj9cc",
     "task": "",
     "save_load_fn": "bitsandbytes",
     "save_load_fn_args": {}