Upload LlamaForCausalLM

config.json

@@ -0,0 +1,42 @@
+{
+  "_name_or_path": "meta-llama-Meta-Llama-3-70B-Instruct-awq",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 128000,
+  "eos_token_id": 128009,
+  "hidden_act": "silu",
+  "hidden_size": 8192,
+  "initializer_range": 0.02,
+  "intermediate_size": 28672,
+  "max_position_embeddings": 8192,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 64,
+  "num_hidden_layers": 80,
+  "num_key_value_heads": 8,
+  "pretraining_tp": 1,
+  "quantization_config": {
+    "backend": "autoawq",
+    "bits": 4,
+    "do_fuse": false,
+    "exllama_config": null,
+    "fuse_max_seq_len": null,
+    "group_size": 128,
+    "modules_to_fuse": null,
+    "modules_to_not_convert": null,
+    "quant_method": "awq",
+    "version": "gemm",
+    "zero_point": true
+  },
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 500000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.41.2",
+  "use_cache": false,
+  "vocab_size": 128256
+}

generation_config.json

@@ -0,0 +1,12 @@
+{
+  "bos_token_id": 128000,
+  "do_sample": true,
+  "eos_token_id": [
+    128001,
+    128009
+  ],
+  "max_length": 4096,
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.41.2"
+}

model-00001-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:541fb369e14a2b96c99954c184ca496a0ce1b5be64ad21696e623b77e9e911bb
+size 4969219128

model-00002-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b95b65a3cd928727ccff4c1c1b5883884220041f250932f6aa66aafbbbb21e6
+size 4890226896

model-00003-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a2daebde0029ba2e52f2f556f18a532da041245f6db580f10da254f8f4d9bb9
+size 4890226992

model-00004-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce536253a29927d86b1eb9317eab42803670213167455d1710b68f67c0f2cfa0
+size 4890226992

model-00005-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a20ac97b86f4b4b7d9f1a263f4bd43e7821ce2394ccb1690168313b943d4894f
+size 4890226992

model-00006-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4c94f301115f509a14f91f9b58edbb413a085813155010bd31d30228afb5a0d
+size 4890226992

model-00007-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44627a425fc5f059a643c96815eab712f546d8b2622e6d60cf749c4b96ceb40f
+size 4890226992

model-00008-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8c06400758aeaa102d153f0bfeecb3d01bfb3301780e56f6cc535f815b658650
+size 3356068840

model-00009-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7f36afb2837cc4b8d6caa0b167f14fc354549c55d6afe96cccea3dd48e84598
+size 2101346432