Update config.json
Browse files- config.json +57 -0
config.json
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "student-t5-base-qa-qg-hl-12-6",
|
| 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 |
+
"init_metadata": {
|
| 13 |
+
"copied_decoder_layers": [
|
| 14 |
+
0,
|
| 15 |
+
2,
|
| 16 |
+
4,
|
| 17 |
+
7,
|
| 18 |
+
9,
|
| 19 |
+
11
|
| 20 |
+
],
|
| 21 |
+
"copied_encoder_layers": [
|
| 22 |
+
0,
|
| 23 |
+
1,
|
| 24 |
+
2,
|
| 25 |
+
3,
|
| 26 |
+
4,
|
| 27 |
+
5,
|
| 28 |
+
6,
|
| 29 |
+
7,
|
| 30 |
+
8,
|
| 31 |
+
9,
|
| 32 |
+
10,
|
| 33 |
+
11
|
| 34 |
+
],
|
| 35 |
+
"teacher_type": "t5"
|
| 36 |
+
},
|
| 37 |
+
"initializer_factor": 1.0,
|
| 38 |
+
"is_encoder_decoder": true,
|
| 39 |
+
"layer_norm_epsilon": 1e-06,
|
| 40 |
+
"model_type": "t5",
|
| 41 |
+
"n_positions": 512,
|
| 42 |
+
"num_decoder_layers": 6,
|
| 43 |
+
"num_heads": 12,
|
| 44 |
+
"num_layers": 12,
|
| 45 |
+
"output_past": true,
|
| 46 |
+
"pad_token_id": 0,
|
| 47 |
+
"relative_attention_num_buckets": 32,
|
| 48 |
+
"task_specific_params": {
|
| 49 |
+
"translation_en_to_fr": {
|
| 50 |
+
"early_stopping": true,
|
| 51 |
+
"max_length": 32,
|
| 52 |
+
"num_beams": 4,
|
| 53 |
+
"prefix": ""
|
| 54 |
+
}
|
| 55 |
+
},
|
| 56 |
+
"vocab_size": 32102
|
| 57 |
+
}
|