Upload MistralForCausalLM

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "unsloth/Mistral-Small-Instruct-2409-bnb-4bit",
+  "_name_or_path": "HobieLLM-merge4",
   "architectures": [
     "MistralForCausalLM"
   ],
@@ -18,7 +18,10 @@
   "num_key_value_heads": 8,
   "pad_token_id": 750,
   "quantization_config": {
+    "_load_in_4bit": true,
+    "_load_in_8bit": false,
     "bnb_4bit_compute_dtype": "bfloat16",
+    "bnb_4bit_quant_storage": "uint8",
     "bnb_4bit_quant_type": "nf4",
     "bnb_4bit_use_double_quant": true,
     "llm_int8_enable_fp32_cpu_offload": false,
@@ -34,7 +37,7 @@
   "rope_theta": 1000000.0,
   "sliding_window": null,
   "tie_word_embeddings": false,
-  "torch_dtype": "bfloat16",
+  "torch_dtype": "float16",
   "transformers_version": "4.44.2",
   "unsloth_version": "2024.9.post4",
   "use_cache": true,

model-00001-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e0cb5938436da9b2db7dca6b99e74ddd95d9fea9352902ecc552c77a27a8768
-size 4979766316
+oid sha256:e0a671f9e5a1ae768412cf25844bd9725f3f4324cba17b38a576962e93692981
+size 4979766219

model-00002-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b35a02483e36215c157d910ea9566c47787728bd5f0e3358ed4b8d6d684a3d0b
-size 4979665814
+oid sha256:5da0d662e51dcc09667e6daf6128c848866635321d69b35a2599f3e89d07fc6f
+size 4979665734

model-00003-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:74385fbb4d73f3f8b71545d82dfdcd51bd601ca74db158ebd0e0e36a89147783
-size 2116663382
+oid sha256:be539c4837dd4f14a7aa2f0555fdb2f915df5224006e40831618103a344d0d17
+size 2116663356

model.safetensors.index.json

@@ -1,6 +1,6 @@
 {
   "metadata": {
-    "total_size": 12075801064
+    "total_size": 12075800989
   },
   "weight_map": {
     "lm_head.weight": "model-00003-of-00003.safetensors",