Update config.json
Browse files- config.json +25 -43
config.json
CHANGED
|
@@ -1,46 +1,28 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
|
| 23 |
-
"attention_layer_norm_rms_epsilon": 0.00001
|
| 24 |
-
}
|
| 25 |
},
|
| 26 |
-
"
|
| 27 |
-
|
| 28 |
-
|
| 29 |
-
|
| 30 |
-
"BOS": 1,
|
| 31 |
-
"EOS": 2,
|
| 32 |
-
"UNK": 0,
|
| 33 |
-
"LF": 13
|
| 34 |
-
}
|
| 35 |
-
},
|
| 36 |
-
"file_metadata": {
|
| 37 |
-
"file_type": "GGUF V2",
|
| 38 |
-
"model_params": "7.24 B",
|
| 39 |
-
"model_size": "4.07 GiB",
|
| 40 |
-
"block_count": 32,
|
| 41 |
-
"embedding_length": 4096,
|
| 42 |
-
"quantization_version": 2,
|
| 43 |
-
"model_path": "Models\\mistral-7b-instruct-v0.2.Q3_K_M.gguf"
|
| 44 |
-
},
|
| 45 |
-
|
| 46 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"architectures": ["MistralForCausalLM"],
|
| 3 |
+
"model_type": "mistral",
|
| 4 |
+
"hidden_size": 4096,
|
| 5 |
+
"intermediate_size": 11008,
|
| 6 |
+
"num_attention_heads": 32,
|
| 7 |
+
"num_hidden_layers": 32,
|
| 8 |
+
"vocab_size": 50272,
|
| 9 |
+
"max_position_embeddings": 2048,
|
| 10 |
+
"rotary_embedding_base": 10000,
|
| 11 |
+
"tie_word_embeddings": false,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"bos_token_id": 1,
|
| 15 |
+
"eos_token_id": 2,
|
| 16 |
+
"pad_token_id": 0,
|
| 17 |
+
"torch_dtype": "float16",
|
| 18 |
+
"use_cache": true,
|
| 19 |
+
"quantization_config": {
|
| 20 |
+
"bits": 4,
|
| 21 |
+
"group_size": 128,
|
| 22 |
+
"dtype": "q4_k_m"
|
|
|
|
|
|
|
| 23 |
},
|
| 24 |
+
"model_revision": "v0.1",
|
| 25 |
+
"instructions": "This model is fine-tuned for instruction-following tasks using the Mistral 7B model.",
|
| 26 |
+
"repository": "Aditya757864/Mistral7B",
|
| 27 |
+
"version": "0.1"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 28 |
}
|