Upload meta-llama/Llama-2-7b-chat-hf ctranslate fp16 weights

README.md

@@ -13,6 +13,7 @@ language:
 - en
 pipeline_tag: text-generation
 inference: false
+arxiv: 2307.09288
 tags:
 - ctranslate2
 - int8
@@ -28,7 +29,7 @@ Speedup inference while reducing memory by 2x-4x using int8 inference in C++ on
 
 quantized version of [meta-llama/Llama-2-7b-chat-hf](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf)
 ```bash
-pip install hf-hub-ctranslate2>=2.12.0 ctranslate2>=3.16.0
+pip install hf-hub-ctranslate2>=2.12.0 ctranslate2>=3.17.1
 ```
 
 ```python
@@ -52,12 +53,12 @@ outputs = model.generate(
 print(outputs)
 ```
 
-Checkpoint compatible to [ctranslate2>=3.16.0](https://github.com/OpenNMT/CTranslate2)
+Checkpoint compatible to [ctranslate2>=3.17.1](https://github.com/OpenNMT/CTranslate2)
 and [hf-hub-ctranslate2>=2.12.0](https://github.com/michaelfeil/hf-hub-ctranslate2)
 - `compute_type=int8_float16` for `device="cuda"`
 - `compute_type=int8`  for `device="cpu"`
 
-Converted on 2023-07-19 using
+Converted on 2023-07-21 using
 ```
 LLama-2 -> removed <pad> token.
 ```
@@ -100,6 +101,8 @@ Meta developed and publicly released the Llama 2 family of large language models
 
 **License** A custom commercial license is available at: [https://ai.meta.com/resources/models-and-libraries/llama-downloads/](https://ai.meta.com/resources/models-and-libraries/llama-downloads/)
 
+**Research Paper** ["Llama-2: Open Foundation and Fine-tuned Chat Models"](arxiv.org/abs/2307.09288)
+
 ## Intended Use
 **Intended Use Cases** Llama 2 is intended for commercial and research use in English. Tuned models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks.
 

config.json

@@ -1,4 +1,5 @@
 {
+    "_name_or_path": "meta-llama/Llama-2-7b-chat-hf",
     "architectures": [
         "LlamaForCausalLM"
     ],
@@ -8,6 +9,7 @@
     "hidden_size": 4096,
     "initializer_range": 0.02,
     "intermediate_size": 11008,
+    "max_length": 4096,
     "max_position_embeddings": 2048,
     "model_type": "llama",
     "num_attention_heads": 32,
@@ -15,15 +17,15 @@
     "num_key_value_heads": 32,
     "pad_token_id": 0,
     "pretraining_tp": 1,
-    "rms_norm_eps": 1e-05,
+    "rms_norm_eps": 1e-06,
     "rope_scaling": null,
     "tie_word_embeddings": false,
     "torch_dtype": "float16",
-    "transformers_version": "4.31.0.dev0",
+    "transformers_version": "4.32.0.dev0",
     "use_cache": true,
     "vocab_size": 32000,
     "bos_token": "<s>",
     "eos_token": "</s>",
-    "layer_norm_epsilon": null,
+    "layer_norm_epsilon": 1e-06,
     "unk_token": "<unk>"
 }

generation_config.json

@@ -2,8 +2,9 @@
   "_from_model_config": true,
   "bos_token_id": 1,
   "eos_token_id": 2,
-  "pad_token_id": 32000,
+  "max_length": 4096,
+  "pad_token_id": 0,
   "temperature": 0.9,
   "top_p": 0.6,
-  "transformers_version": "4.31.0.dev0"
+  "transformers_version": "4.32.0.dev0"
 }

special_tokens_map.json

@@ -2,22 +2,21 @@
   "bos_token": {
     "content": "<s>",
     "lstrip": false,
-    "normalized": true,
+    "normalized": false,
     "rstrip": false,
     "single_word": false
   },
   "eos_token": {
     "content": "</s>",
     "lstrip": false,
-    "normalized": true,
+    "normalized": false,
     "rstrip": false,
     "single_word": false
   },
-  "pad_token": "<unk>",
   "unk_token": {
     "content": "<unk>",
     "lstrip": false,
-    "normalized": true,
+    "normalized": false,
     "rstrip": false,
     "single_word": false
   }

tokenizer.json

@@ -9,7 +9,7 @@
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
-      "normalized": true,
+      "normalized": false,
       "special": true
     },
     {
@@ -18,7 +18,7 @@
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
-      "normalized": true,
+      "normalized": false,
       "special": true
     },
     {
@@ -27,7 +27,7 @@
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
-      "normalized": true,
+      "normalized": false,
       "special": true
     }
   ],

tokenizer_config.json

@@ -5,7 +5,7 @@
     "__type": "AddedToken",
     "content": "<s>",
     "lstrip": false,
-    "normalized": true,
+    "normalized": false,
     "rstrip": false,
     "single_word": false
   },
@@ -14,7 +14,7 @@
     "__type": "AddedToken",
     "content": "</s>",
     "lstrip": false,
-    "normalized": true,
+    "normalized": false,
     "rstrip": false,
     "single_word": false
   },
@@ -27,8 +27,9 @@
     "__type": "AddedToken",
     "content": "<unk>",
     "lstrip": false,
-    "normalized": true,
+    "normalized": false,
     "rstrip": false,
     "single_word": false
-  }
+  },
+  "use_fast": false
 }