Update config.json
Browse files- config.json +8 -68
config.json
CHANGED
|
@@ -1,96 +1,36 @@
|
|
| 1 |
{
|
| 2 |
-
|
| 3 |
"architectures": [
|
| 4 |
-
|
| 5 |
"T5ForConditionalGeneration"
|
| 6 |
-
|
| 7 |
],
|
| 8 |
-
|
| 9 |
"d_ff": 3072,
|
| 10 |
-
|
| 11 |
"d_kv": 64,
|
| 12 |
-
|
| 13 |
"d_model": 768,
|
| 14 |
-
|
| 15 |
"decoder_start_token_id": 0,
|
| 16 |
-
|
| 17 |
"dropout_rate": 0.1,
|
| 18 |
-
|
| 19 |
"eos_token_id": 1,
|
| 20 |
-
|
| 21 |
"initializer_factor": 1.0,
|
| 22 |
-
|
| 23 |
"is_encoder_decoder": true,
|
| 24 |
-
|
| 25 |
"layer_norm_epsilon": 1e-06,
|
| 26 |
-
|
| 27 |
"model_type": "t5",
|
| 28 |
-
|
| 29 |
"n_positions": 512,
|
| 30 |
-
|
| 31 |
"num_heads": 12,
|
| 32 |
-
|
| 33 |
"num_layers": 12,
|
| 34 |
-
|
| 35 |
"output_past": true,
|
| 36 |
-
|
| 37 |
"pad_token_id": 0,
|
| 38 |
-
|
| 39 |
"relative_attention_num_buckets": 32,
|
| 40 |
-
|
| 41 |
"task_specific_params": {
|
| 42 |
-
|
| 43 |
-
"summarization": {
|
| 44 |
-
|
| 45 |
-
"early_stopping": true,
|
| 46 |
-
|
| 47 |
-
"max_length": 32,
|
| 48 |
-
|
| 49 |
-
"num_beams": 4,
|
| 50 |
-
|
| 51 |
-
"prefix": ""
|
| 52 |
-
|
| 53 |
-
},
|
| 54 |
-
|
| 55 |
"translation_en_to_fr": {
|
| 56 |
-
|
| 57 |
-
"early_stopping": true,
|
| 58 |
-
|
| 59 |
-
"max_length": 32,
|
| 60 |
-
|
| 61 |
-
"num_beams": 4,
|
| 62 |
-
|
| 63 |
-
"prefix": ""
|
| 64 |
-
|
| 65 |
-
},
|
| 66 |
-
|
| 67 |
-
"translation_en_to_de": {
|
| 68 |
-
|
| 69 |
-
"early_stopping": true,
|
| 70 |
-
|
| 71 |
-
"max_length": 32,
|
| 72 |
-
|
| 73 |
-
"num_beams": 4,
|
| 74 |
-
|
| 75 |
-
"prefix": ""
|
| 76 |
-
|
| 77 |
-
},
|
| 78 |
-
|
| 79 |
-
"translation_en_to_ro": {
|
| 80 |
-
|
| 81 |
"early_stopping": true,
|
| 82 |
-
|
| 83 |
"max_length": 32,
|
| 84 |
-
|
| 85 |
"num_beams": 4,
|
| 86 |
-
|
| 87 |
"prefix": ""
|
| 88 |
-
|
| 89 |
}
|
| 90 |
-
|
| 91 |
},
|
| 92 |
-
|
| 93 |
-
"
|
| 94 |
-
|
| 95 |
-
|
| 96 |
-
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "ck46/t5-base-hotpot-qa-qg",
|
| 3 |
"architectures": [
|
|
|
|
| 4 |
"T5ForConditionalGeneration"
|
|
|
|
| 5 |
],
|
|
|
|
| 6 |
"d_ff": 3072,
|
|
|
|
| 7 |
"d_kv": 64,
|
|
|
|
| 8 |
"d_model": 768,
|
|
|
|
| 9 |
"decoder_start_token_id": 0,
|
|
|
|
| 10 |
"dropout_rate": 0.1,
|
|
|
|
| 11 |
"eos_token_id": 1,
|
| 12 |
+
"feed_forward_proj": "relu",
|
| 13 |
"initializer_factor": 1.0,
|
|
|
|
| 14 |
"is_encoder_decoder": true,
|
|
|
|
| 15 |
"layer_norm_epsilon": 1e-06,
|
|
|
|
| 16 |
"model_type": "t5",
|
|
|
|
| 17 |
"n_positions": 512,
|
| 18 |
+
"num_decoder_layers": 12,
|
| 19 |
"num_heads": 12,
|
|
|
|
| 20 |
"num_layers": 12,
|
|
|
|
| 21 |
"output_past": true,
|
|
|
|
| 22 |
"pad_token_id": 0,
|
|
|
|
| 23 |
"relative_attention_num_buckets": 32,
|
|
|
|
| 24 |
"task_specific_params": {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 25 |
"translation_en_to_fr": {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 26 |
"early_stopping": true,
|
|
|
|
| 27 |
"max_length": 32,
|
|
|
|
| 28 |
"num_beams": 4,
|
|
|
|
| 29 |
"prefix": ""
|
|
|
|
| 30 |
}
|
|
|
|
| 31 |
},
|
| 32 |
+
"torch_dtype": "float32",
|
| 33 |
+
"transformers_version": "4.15.0",
|
| 34 |
+
"use_cache": true,
|
| 35 |
+
"vocab_size": 32102
|
| 36 |
+
}
|