Upload LlamaForCausalLM

config.json

@@ -0,0 +1,42 @@
+{
+  "_name_or_path": "./dolphin-2.9.1-llama-3-70b-awq",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 128000,
+  "eos_token_id": 128256,
+  "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": "4.41.0",
+  "use_cache": false,
+  "vocab_size": 128258
+}

generation_config.json

@@ -0,0 +1,9 @@
+{
+  "bos_token_id": 128000,
+  "do_sample": true,
+  "eos_token_id": 128001,
+  "max_length": 4096,
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.41.0"
+}

model-00001-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b3018fcce46ffdf0f1775f6c792cc2f0ab414442e61271fe7e70c04c4d98e380
+size 4969251896

model-00002-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc601b43b4e24c2bebb3db3ca8e2c9f23a2412e8afbb661e8e540afd761142d5
+size 4890226896

model-00003-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87dc8765d64d2325d6a2df1d1e47bad938d0be6bdef69f4c0768ba66854ce245
+size 4890226992

model-00004-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f68d38a7d5b48f306c0f4633983653711e33ebab7c99639d204d6b542873cf7
+size 4890226992

model-00005-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:205817ac63732c0b444e23c8e2be6172c206fd821a0a7c32f8181fd70340998b
+size 4890226992

model-00006-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b6c7cabca45b21723ba6fac935efc28e0ab1fbf842fb782332fe57a679a8595
+size 4890226992

model-00007-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:096b8225689190f05474cb40f406166049c389064360f753e921369e6ff44d79
+size 4890226992

model-00008-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:704089acb46967bd850f80ecca0ef681f43ab0bef047b8821627974b2f745282
+size 3356068840

model-00009-of-00009.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56d691dce48b6d96cffc8f8d2b4ae3421ce64ff46865c976db0bb0b718dd9b47
+size 2101379200