updated to mistral 7b base 0.2 from 0.1

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "models/koishi-mistral/merged/",
+  "_name_or_path": "models/koishi-7b-qlora-lora/merged/",
   "architectures": [
     "MistralForCausalLM"
   ],
@@ -15,12 +15,25 @@
   "num_attention_heads": 32,
   "num_hidden_layers": 32,
   "num_key_value_heads": 8,
+  "quantization_config": {
+    "bits": 4,
+    "damp_percent": 0.1,
+    "desc_act": true,
+    "group_size": -1,
+    "is_marlin_format": false,
+    "model_file_base_name": null,
+    "model_name_or_path": null,
+    "quant_method": "gptq",
+    "static_groups": false,
+    "sym": true,
+    "true_sequential": true
+  },
   "rms_norm_eps": 1e-05,
-  "rope_theta": 10000.0,
-  "sliding_window": 4096,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
   "tie_word_embeddings": false,
   "torch_dtype": "float16",
-  "transformers_version": "4.36.0.dev0",
+  "transformers_version": "4.40.0.dev0",
   "use_cache": false,
   "vocab_size": 32003
 }

gptq_model-4bit--1g.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f1f3ae9c1422c28d6a221a065c7d1861472b3c8af745e7e92e9752ceebfea2b
+size 4025836640

quantize_config.json

@@ -1,11 +1,13 @@
 {
   "bits": 4,
-  "group_size": 128,
+  "group_size": -1,
   "damp_percent": 0.1,
   "desc_act": true,
   "static_groups": false,
   "sym": true,
   "true_sequential": true,
   "model_name_or_path": null,
-  "model_file_base_name": null
+  "model_file_base_name": null,
+  "is_marlin_format": false,
+  "quant_method": "gptq"
 }

tokenizer_config.json

@@ -1,6 +1,7 @@
 {
   "add_bos_token": true,
   "add_eos_token": false,
+  "add_prefix_space": true,
   "added_tokens_decoder": {
     "0": {
       "content": "<unk>",
@@ -51,7 +52,6 @@
       "special": false
     }
   },
-  "additional_special_tokens": [],
   "bos_token": "<s>",
   "clean_up_tokenization_spaces": false,
   "eos_token": "</s>",
@@ -61,7 +61,6 @@
   "sp_model_kwargs": {},
   "spaces_between_special_tokens": false,
   "tokenizer_class": "LlamaTokenizer",
-  "trust_remote_code": true,
   "unk_token": "<unk>",
   "use_default_system_prompt": false,
   "use_fast": true