uploaded model
Browse files- config.json +185 -0
config.json
ADDED
|
@@ -0,0 +1,185 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Phi-3-medium-128k-instruct",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"Phi3ForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"auto_map": {
|
| 9 |
+
"AutoConfig": "configuration_phi3.Phi3Config",
|
| 10 |
+
"AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM"
|
| 11 |
+
},
|
| 12 |
+
"bos_token_id": 1,
|
| 13 |
+
"embd_pdrop": 0.0,
|
| 14 |
+
"eos_token_id": 32000,
|
| 15 |
+
"hidden_act": "silu",
|
| 16 |
+
"hidden_size": 5120,
|
| 17 |
+
"initializer_range": 0.02,
|
| 18 |
+
"intermediate_size": 17920,
|
| 19 |
+
"max_position_embeddings": 131072,
|
| 20 |
+
"model_type": "phi3",
|
| 21 |
+
"num_attention_heads": 40,
|
| 22 |
+
"num_hidden_layers": 40,
|
| 23 |
+
"num_key_value_heads": 10,
|
| 24 |
+
"original_max_position_embeddings": 4096,
|
| 25 |
+
"pad_token_id": null,
|
| 26 |
+
"quantization_config": {
|
| 27 |
+
"_load_in_4bit": true,
|
| 28 |
+
"_load_in_8bit": false,
|
| 29 |
+
"bnb_4bit_compute_dtype": "float32",
|
| 30 |
+
"bnb_4bit_quant_storage": "uint8",
|
| 31 |
+
"bnb_4bit_quant_type": "fp4",
|
| 32 |
+
"bnb_4bit_use_double_quant": false,
|
| 33 |
+
"llm_int8_enable_fp32_cpu_offload": false,
|
| 34 |
+
"llm_int8_has_fp16_weight": false,
|
| 35 |
+
"llm_int8_skip_modules": null,
|
| 36 |
+
"llm_int8_threshold": 6.0,
|
| 37 |
+
"load_in_4bit": true,
|
| 38 |
+
"load_in_8bit": false,
|
| 39 |
+
"quant_method": "bitsandbytes"
|
| 40 |
+
},
|
| 41 |
+
"resid_pdrop": 0.0,
|
| 42 |
+
"rms_norm_eps": 1e-05,
|
| 43 |
+
"rope_scaling": {
|
| 44 |
+
"long_factor": [
|
| 45 |
+
1.0,
|
| 46 |
+
1.0,
|
| 47 |
+
1.0,
|
| 48 |
+
1.0,
|
| 49 |
+
1.0,
|
| 50 |
+
1.0,
|
| 51 |
+
1.0,
|
| 52 |
+
1.0,
|
| 53 |
+
1.0,
|
| 54 |
+
1.0,
|
| 55 |
+
1.0,
|
| 56 |
+
1.0,
|
| 57 |
+
1.0,
|
| 58 |
+
1.25,
|
| 59 |
+
1.25,
|
| 60 |
+
1.5,
|
| 61 |
+
2.0,
|
| 62 |
+
2.75,
|
| 63 |
+
5.75,
|
| 64 |
+
5.75,
|
| 65 |
+
6.5,
|
| 66 |
+
9.25,
|
| 67 |
+
11.0,
|
| 68 |
+
13.25,
|
| 69 |
+
19.25,
|
| 70 |
+
19.75,
|
| 71 |
+
19.75,
|
| 72 |
+
21.25,
|
| 73 |
+
21.5,
|
| 74 |
+
26.5,
|
| 75 |
+
30.0,
|
| 76 |
+
33.75,
|
| 77 |
+
35.25,
|
| 78 |
+
38.5,
|
| 79 |
+
42.0,
|
| 80 |
+
42.25,
|
| 81 |
+
46.0,
|
| 82 |
+
47.0,
|
| 83 |
+
50.0,
|
| 84 |
+
50.5,
|
| 85 |
+
51.0,
|
| 86 |
+
52.0,
|
| 87 |
+
52.75,
|
| 88 |
+
53.75,
|
| 89 |
+
54.75,
|
| 90 |
+
57.0,
|
| 91 |
+
57.25,
|
| 92 |
+
58.5,
|
| 93 |
+
59.25,
|
| 94 |
+
59.5,
|
| 95 |
+
62.0,
|
| 96 |
+
62.5,
|
| 97 |
+
62.75,
|
| 98 |
+
63.25,
|
| 99 |
+
63.25,
|
| 100 |
+
63.25,
|
| 101 |
+
63.75,
|
| 102 |
+
64.0,
|
| 103 |
+
64.0,
|
| 104 |
+
64.25,
|
| 105 |
+
64.5,
|
| 106 |
+
64.5,
|
| 107 |
+
65.0,
|
| 108 |
+
65.0
|
| 109 |
+
],
|
| 110 |
+
"short_factor": [
|
| 111 |
+
1.0,
|
| 112 |
+
1.0,
|
| 113 |
+
1.0,
|
| 114 |
+
1.0,
|
| 115 |
+
1.0,
|
| 116 |
+
1.0,
|
| 117 |
+
1.01,
|
| 118 |
+
1.02,
|
| 119 |
+
1.02,
|
| 120 |
+
1.04,
|
| 121 |
+
1.04,
|
| 122 |
+
1.07,
|
| 123 |
+
1.07,
|
| 124 |
+
1.1,
|
| 125 |
+
1.3000000000000003,
|
| 126 |
+
1.3000000000000003,
|
| 127 |
+
1.5000000000000004,
|
| 128 |
+
1.5700000000000005,
|
| 129 |
+
1.9000000000000008,
|
| 130 |
+
2.3100000000000014,
|
| 131 |
+
2.759999999999992,
|
| 132 |
+
3.3899999999999784,
|
| 133 |
+
3.9399999999999666,
|
| 134 |
+
4.009999999999965,
|
| 135 |
+
4.289999999999959,
|
| 136 |
+
4.349999999999958,
|
| 137 |
+
5.349999999999937,
|
| 138 |
+
6.659999999999909,
|
| 139 |
+
7.029999999999901,
|
| 140 |
+
7.51999999999989,
|
| 141 |
+
8.00999999999988,
|
| 142 |
+
8.249999999999876,
|
| 143 |
+
8.279999999999875,
|
| 144 |
+
9.629999999999846,
|
| 145 |
+
9.89999999999984,
|
| 146 |
+
10.589999999999826,
|
| 147 |
+
11.049999999999816,
|
| 148 |
+
11.7899999999998,
|
| 149 |
+
12.189999999999792,
|
| 150 |
+
12.889999999999777,
|
| 151 |
+
13.129999999999772,
|
| 152 |
+
13.16999999999977,
|
| 153 |
+
13.20999999999977,
|
| 154 |
+
13.479999999999764,
|
| 155 |
+
13.539999999999763,
|
| 156 |
+
13.779999999999758,
|
| 157 |
+
13.929999999999755,
|
| 158 |
+
14.429999999999744,
|
| 159 |
+
14.759999999999737,
|
| 160 |
+
15.149999999999729,
|
| 161 |
+
15.419999999999723,
|
| 162 |
+
15.53999999999972,
|
| 163 |
+
15.659999999999718,
|
| 164 |
+
15.749999999999716,
|
| 165 |
+
15.759999999999716,
|
| 166 |
+
15.799999999999715,
|
| 167 |
+
16.05999999999971,
|
| 168 |
+
16.079999999999714,
|
| 169 |
+
16.11999999999972,
|
| 170 |
+
16.11999999999972,
|
| 171 |
+
16.18999999999973,
|
| 172 |
+
16.31999999999975,
|
| 173 |
+
16.539999999999786,
|
| 174 |
+
16.799999999999827
|
| 175 |
+
],
|
| 176 |
+
"type": "longrope"
|
| 177 |
+
},
|
| 178 |
+
"rope_theta": 10000.0,
|
| 179 |
+
"sliding_window": 131072,
|
| 180 |
+
"tie_word_embeddings": false,
|
| 181 |
+
"torch_dtype": "float16",
|
| 182 |
+
"transformers_version": "4.44.2",
|
| 183 |
+
"use_cache": true,
|
| 184 |
+
"vocab_size": 32064
|
| 185 |
+
}
|