AWQ model commit

added_tokens.json

@@ -0,0 +1,7 @@
+{
+  "</s>": 2,
+  "<s>": 1,
+  "<unk>": 0,
+  "<|im_end|>": 32000,
+  "<|im_start|>": 32001
+}

config.json

@@ -0,0 +1,42 @@
+{
+  "_name_or_path": "/workspace/process/paixai_astrid-mistral-7b/source",
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "bos_token_id": 1,
+  "custom_pipelines": {
+    "text-generation": {
+      "impl": "h2oai_pipeline.H2OTextGenerationPipeline",
+      "pt": "AutoModelForCausalLM"
+    }
+  },
+  "eos_token_id": 32000,
+  "hidden_act": "silu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 32768,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "pad_token_id": 0,
+  "pretraining_tp": 1,
+  "quantization_config": {
+    "bits": 4,
+    "group_size": 128,
+    "quant_method": "awq",
+    "version": "gemm",
+    "zero_point": true
+  },
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 10000.0,
+  "sliding_window": 4096,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.35.2",
+  "use_cache": true,
+  "vocab_size": 32002
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 32000,
+  "pad_token_id": 0,
+  "transformers_version": "4.34.0"
+}

h2oai_pipeline.py

@@ -0,0 +1,42 @@
+from transformers import TextGenerationPipeline
+from transformers.pipelines.text_generation import ReturnType
+
+STYLE = "<|prompt|>{instruction}<|im_end|><|answer|>"
+
+
+class H2OTextGenerationPipeline(TextGenerationPipeline):
+    def __init__(self, *args, **kwargs):
+        super().__init__(*args, **kwargs)
+        self.prompt = STYLE
+
+    def preprocess(
+        self, prompt_text, prefix="", handle_long_generation=None, **generate_kwargs
+    ):
+        prompt_text = self.prompt.format(instruction=prompt_text)
+        return super().preprocess(
+            prompt_text,
+            prefix=prefix,
+            handle_long_generation=handle_long_generation,
+            **generate_kwargs,
+        )
+
+    def postprocess(
+        self,
+        model_outputs,
+        return_type=ReturnType.FULL_TEXT,
+        clean_up_tokenization_spaces=True,
+    ):
+        records = super().postprocess(
+            model_outputs,
+            return_type=return_type,
+            clean_up_tokenization_spaces=clean_up_tokenization_spaces,
+        )
+        for rec in records:
+            rec["generated_text"] = (
+                rec["generated_text"]
+                .split("<|answer|>")[1]
+                .strip()
+                .split("<|prompt|>")[0]
+                .strip()
+            )
+        return records

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88775ebab9e344547fafb4dc8745a20a391bec1fa750f28010be00ce798f3e62
+size 4150913000

quant_config.json

@@ -0,0 +1,6 @@
+{
+    "zero_point": true,
+    "q_group_size": 128,
+    "w_bit": 4,
+    "version": "GEMM"
+}

special_tokens_map.json

@@ -0,0 +1,15 @@
+{
+  "additional_special_tokens": [
+    "<unk>",
+    "<s>",
+    "</s>",
+    "<|im_end|>",
+    "<|im_start|>"
+  ],
+  "bos_token": "<s>",
+  "cls_token": "<|im_end|>",
+  "eos_token": "<|im_end|>",
+  "pad_token": "<unk>",
+  "sep_token": "<|im_end|>",
+  "unk_token": "<unk>"
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443

tokenizer_config.json

@@ -0,0 +1,68 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32000": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32001": {
+      "content": "<|im_start|>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<unk>",
+    "<s>",
+    "</s>",
+    "<|im_end|>",
+    "<|im_start|>"
+  ],
+  "bos_token": "<s>",
+  "chat_template": "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "legacy": true,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<unk>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "trust_remote_code": false,
+  "unk_token": "<unk>",
+  "use_default_system_prompt": true,
+  "use_fast": true
+}