Update config.json
Browse files- config.json +113 -15
config.json
CHANGED
|
@@ -1,36 +1,134 @@
|
|
| 1 |
-
{
|
| 2 |
-
"_name_or_path": "sciphi/triplex",
|
| 3 |
"architectures": [
|
| 4 |
-
"
|
| 5 |
],
|
|
|
|
| 6 |
"attention_dropout": 0.0,
|
| 7 |
"auto_map": {
|
| 8 |
-
"AutoConfig": "configuration_phi3.Phi3Config",
|
| 9 |
-
"
|
|
|
|
| 10 |
},
|
| 11 |
"bos_token_id": 1,
|
| 12 |
"embd_pdrop": 0.0,
|
| 13 |
-
"eos_token_id": 32000,
|
| 14 |
-
"hidden_act": "silu",
|
| 15 |
"hidden_size": 3072,
|
| 16 |
"initializer_range": 0.02,
|
| 17 |
"intermediate_size": 8192,
|
| 18 |
-
"max_position_embeddings":
|
| 19 |
"model_type": "phi3",
|
| 20 |
"num_attention_heads": 32,
|
| 21 |
"num_hidden_layers": 32,
|
| 22 |
-
"num_key_value_heads": 32,
|
| 23 |
-
"original_max_position_embeddings": 4096,
|
| 24 |
"pad_token_id": 32000,
|
| 25 |
"resid_pdrop": 0.0,
|
| 26 |
"rms_norm_eps": 1e-05,
|
| 27 |
-
"rope_scaling":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 28 |
"rope_theta": 10000.0,
|
| 29 |
-
"sliding_window":
|
| 30 |
"tie_word_embeddings": false,
|
| 31 |
"torch_dtype": "bfloat16",
|
| 32 |
-
"transformers_version": "4.
|
| 33 |
-
"use_cache":
|
| 34 |
-
|
| 35 |
"vocab_size": 32064
|
| 36 |
}
|
|
|
|
|
|
|
|
|
|
| 1 |
"architectures": [
|
| 2 |
+
"sciphi/triplex"
|
| 3 |
],
|
| 4 |
+
"attention_bias": false,
|
| 5 |
"attention_dropout": 0.0,
|
| 6 |
"auto_map": {
|
| 7 |
+
"AutoConfig": "microsoft/Phi-3-mini-128k-instruct--configuration_phi3.Phi3Config",
|
| 8 |
+
"AutoModel": "microsoft/Phi-3-mini-128k-instruct--modeling_phi3.Phi3ForCausalLM",
|
| 9 |
+
"AutoModelForCausalLM": "microsoft/Phi-3-mini-128k-instruct--modeling_phi3.Phi3ForCausalLM"
|
| 10 |
},
|
| 11 |
"bos_token_id": 1,
|
| 12 |
"embd_pdrop": 0.0,
|
|
|
|
|
|
|
| 13 |
"hidden_size": 3072,
|
| 14 |
"initializer_range": 0.02,
|
| 15 |
"intermediate_size": 8192,
|
| 16 |
+
"max_position_embeddings": 131072,
|
| 17 |
"model_type": "phi3",
|
| 18 |
"num_attention_heads": 32,
|
| 19 |
"num_hidden_layers": 32,
|
|
|
|
|
|
|
| 20 |
"pad_token_id": 32000,
|
| 21 |
"resid_pdrop": 0.0,
|
| 22 |
"rms_norm_eps": 1e-05,
|
| 23 |
+
"rope_scaling": {
|
| 24 |
+
"long_factor": [
|
| 25 |
+
1.0700000524520874,
|
| 26 |
+
1.1200000047683716,
|
| 27 |
+
1.149999976158142,
|
| 28 |
+
1.4199999570846558,
|
| 29 |
+
1.5699999332427979,
|
| 30 |
+
1.7999999523162842,
|
| 31 |
+
2.129999876022339,
|
| 32 |
+
2.129999876022339,
|
| 33 |
+
3.009999990463257,
|
| 34 |
+
5.910000324249268,
|
| 35 |
+
6.950000286102295,
|
| 36 |
+
9.070000648498535,
|
| 37 |
+
9.930000305175781,
|
| 38 |
+
10.710000038146973,
|
| 39 |
+
11.130000114440918,
|
| 40 |
+
14.609999656677246,
|
| 41 |
+
15.409998893737793,
|
| 42 |
+
19.809999465942383,
|
| 43 |
+
37.279998779296875,
|
| 44 |
+
38.279998779296875,
|
| 45 |
+
38.599998474121094,
|
| 46 |
+
40.12000274658203,
|
| 47 |
+
46.20000457763672,
|
| 48 |
+
50.940006256103516,
|
| 49 |
+
53.66000747680664,
|
| 50 |
+
54.9373893737793,
|
| 51 |
+
56.89738845825195,
|
| 52 |
+
57.28738784790039,
|
| 53 |
+
59.98738479614258,
|
| 54 |
+
60.86738586425781,
|
| 55 |
+
60.887386322021484,
|
| 56 |
+
61.71739196777344,
|
| 57 |
+
62.91739273071289,
|
| 58 |
+
62.957393646240234,
|
| 59 |
+
63.41739273071289,
|
| 60 |
+
63.8173942565918,
|
| 61 |
+
63.83739471435547,
|
| 62 |
+
63.897396087646484,
|
| 63 |
+
63.93739700317383,
|
| 64 |
+
64.06739807128906,
|
| 65 |
+
64.11434936523438,
|
| 66 |
+
64.12435150146484,
|
| 67 |
+
64.15435028076172,
|
| 68 |
+
64.19435119628906,
|
| 69 |
+
64.24435424804688,
|
| 70 |
+
64.57435607910156,
|
| 71 |
+
64.69000244140625,
|
| 72 |
+
64.76000213623047
|
| 73 |
+
],
|
| 74 |
+
"short_factor": [
|
| 75 |
+
1.1,
|
| 76 |
+
1.1,
|
| 77 |
+
1.1,
|
| 78 |
+
1.3000000000000003,
|
| 79 |
+
1.3500000000000003,
|
| 80 |
+
1.3500000000000003,
|
| 81 |
+
1.4000000000000004,
|
| 82 |
+
1.5500000000000005,
|
| 83 |
+
2.000000000000001,
|
| 84 |
+
2.000000000000001,
|
| 85 |
+
2.000000000000001,
|
| 86 |
+
2.000000000000001,
|
| 87 |
+
2.000000000000001,
|
| 88 |
+
2.000000000000001,
|
| 89 |
+
2.000000000000001,
|
| 90 |
+
2.000000000000001,
|
| 91 |
+
2.000000000000001,
|
| 92 |
+
2.000000000000001,
|
| 93 |
+
2.000000000000001,
|
| 94 |
+
2.000000000000001,
|
| 95 |
+
2.000000000000001,
|
| 96 |
+
2.000000000000001,
|
| 97 |
+
2.000000000000001,
|
| 98 |
+
2.000000000000001,
|
| 99 |
+
2.000000000000001,
|
| 100 |
+
2.0500000000000007,
|
| 101 |
+
2.0500000000000007,
|
| 102 |
+
2.0500000000000007,
|
| 103 |
+
2.0500000000000007,
|
| 104 |
+
2.0500000000000007,
|
| 105 |
+
2.0500000000000007,
|
| 106 |
+
2.1000000000000005,
|
| 107 |
+
2.1000000000000005,
|
| 108 |
+
2.1500000000000004,
|
| 109 |
+
2.25,
|
| 110 |
+
2.25,
|
| 111 |
+
2.25,
|
| 112 |
+
2.25,
|
| 113 |
+
2.25,
|
| 114 |
+
2.3999999999999995,
|
| 115 |
+
2.4499999999999993,
|
| 116 |
+
2.499999999999999,
|
| 117 |
+
2.6999999999999984,
|
| 118 |
+
2.6999999999999984,
|
| 119 |
+
2.7499999999999982,
|
| 120 |
+
2.799999999999998,
|
| 121 |
+
2.8999999999999977,
|
| 122 |
+
3.049999999999997
|
| 123 |
+
],
|
| 124 |
+
"type": "yarn"
|
| 125 |
+
},
|
| 126 |
"rope_theta": 10000.0,
|
| 127 |
+
"sliding_window": 262144,
|
| 128 |
"tie_word_embeddings": false,
|
| 129 |
"torch_dtype": "bfloat16",
|
| 130 |
+
"transformers_version": "4.42.3",
|
| 131 |
+
"use_cache": false,
|
| 132 |
+
|
| 133 |
"vocab_size": 32064
|
| 134 |
}
|