🚀 Subida de modelo cuantizado en int8 con Optimum

config.json

@@ -0,0 +1,42 @@
+{
+  "_attn_implementation_autoset": true,
+  "_name_or_path": "hedtorresca/llama3-1b-lora-fusionado",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "backend": "default",
+  "bos_token_id": 128000,
+  "eos_token_id": [
+    128001,
+    128008,
+    128009
+  ],
+  "head_dim": 64,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "max_position_embeddings": 131072,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 16,
+  "num_key_value_heads": 8,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": {
+    "factor": 32.0,
+    "high_freq_factor": 4.0,
+    "low_freq_factor": 1.0,
+    "original_max_position_embeddings": 8192,
+    "rope_type": "llama3"
+  },
+  "rope_theta": 500000.0,
+  "tie_word_embeddings": true,
+  "torch_dtype": "int8",
+  "transformers_version": "4.48.3",
+  "use_cache": true,
+  "vocab_size": 128256
+}

inc_config.json

@@ -0,0 +1,12 @@
+{
+  "distillation": {},
+  "neural_compressor_version": "3.3.1",
+  "optimum_version": "1.24.0",
+  "pruning": {},
+  "quantization": {
+    "dataset_num_samples": null,
+    "is_static": false
+  },
+  "torch_version": "2.6.0+cu124",
+  "transformers_version": "4.48.3"
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6781a4123bf2632bb39543bad86b0e7ae66668b12535c2008735176e80ceb16c
+size 1508033518