Trained with Unsloth

README.md

@@ -5,4 +5,6 @@ license: apache-2.0
 tags:
 - code
 - unsloth
+- trl
+- sft
 ---

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "outputs",
+  "_name_or_path": "shenzhi-wang/Llama3-8B-Chinese-Chat",
   "architectures": [
     "LlamaForCausalLM"
   ],
@@ -19,18 +19,6 @@
   "num_key_value_heads": 8,
   "pad_token_id": 128255,
   "pretraining_tp": 1,
-  "quantization_config": {
-    "bnb_4bit_compute_dtype": "float16",
-    "bnb_4bit_quant_type": "nf4",
-    "bnb_4bit_use_double_quant": true,
-    "llm_int8_enable_fp32_cpu_offload": false,
-    "llm_int8_has_fp16_weight": false,
-    "llm_int8_skip_modules": null,
-    "llm_int8_threshold": 6.0,
-    "load_in_4bit": true,
-    "load_in_8bit": false,
-    "quant_method": "bitsandbytes"
-  },
   "rms_norm_eps": 1e-05,
   "rope_scaling": null,
   "rope_theta": 500000.0,

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:450e03168edd713ba26fc55667d4c6e95ce45adbabe53aeeb0293c6be80e7766
+size 4976698592

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b65634044af920338a1d0e2e5b1255be0ab6d9ef12573194ceb997cf799d3d1e
+size 4999802616

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07491690af76ba21b48a4aff2603c46d2f80fd8874e00b0b51d203f5b98da38b
+size 4915916080

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2cc89e9ef38bd3d37629321bc2e5d2e895964bce1b76d87273d4ecae44e594e
+size 1168138808