Commit
•
3dcd7a3
1
Parent(s):
3b6576e
Upload MT5ForConditionalGeneration
Browse files- config.json +37 -0
- pytorch_model.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/data-big-projects/danish-summarization-danewsroom/models/chocolate-haze-288/checkpoint-129080",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"MT5ForConditionalGeneration"
|
| 5 |
+
],
|
| 6 |
+
"d_ff": 2048,
|
| 7 |
+
"d_kv": 64,
|
| 8 |
+
"d_model": 768,
|
| 9 |
+
"decoder_start_token_id": 0,
|
| 10 |
+
"dense_act_fn": "gelu_new",
|
| 11 |
+
"dropout_rate": 0.1,
|
| 12 |
+
"eos_token_id": 1,
|
| 13 |
+
"feed_forward_proj": "gated-gelu",
|
| 14 |
+
"initializer_factor": 1.0,
|
| 15 |
+
"is_encoder_decoder": true,
|
| 16 |
+
"is_gated_act": true,
|
| 17 |
+
"layer_norm_epsilon": 1e-06,
|
| 18 |
+
"length_penalty": 5.0,
|
| 19 |
+
"max_length": 128,
|
| 20 |
+
"min_length": 9,
|
| 21 |
+
"model_type": "mt5",
|
| 22 |
+
"no_repeat_ngram_size": 3,
|
| 23 |
+
"num_beams": 2,
|
| 24 |
+
"num_decoder_layers": 12,
|
| 25 |
+
"num_heads": 12,
|
| 26 |
+
"num_layers": 12,
|
| 27 |
+
"output_past": true,
|
| 28 |
+
"pad_token_id": 0,
|
| 29 |
+
"relative_attention_max_distance": 128,
|
| 30 |
+
"relative_attention_num_buckets": 32,
|
| 31 |
+
"tie_word_embeddings": false,
|
| 32 |
+
"tokenizer_class": "T5Tokenizer",
|
| 33 |
+
"torch_dtype": "float32",
|
| 34 |
+
"transformers_version": "4.22.0.dev0",
|
| 35 |
+
"use_cache": true,
|
| 36 |
+
"vocab_size": 250112
|
| 37 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70bcc562817484a8f78559a22849dc5dec01b40462df496da4b40ea9cb0b0742
|
| 3 |
+
size 2329696205
|