Sayan01 commited on
Commit
b4e603f
·
1 Parent(s): 699acb7

Upload T5ForConditionalGeneration

Browse files
Files changed (2) hide show
  1. config.json +6 -5
  2. pytorch_model.bin +2 -2
config.json CHANGED
@@ -1,20 +1,20 @@
1
  {
2
- "_name_or_path": "t5-base",
3
  "architectures": [
4
  "T5ForConditionalGeneration"
5
  ],
6
  "classifier_dropout": 0.0,
7
- "d_ff": 3072,
8
  "d_kv": 64,
9
  "d_model": 768,
10
  "decoder_start_token_id": 0,
11
- "dense_act_fn": "relu",
12
  "dropout_rate": 0.1,
13
  "eos_token_id": 1,
14
- "feed_forward_proj": "relu",
15
  "initializer_factor": 1.0,
16
  "is_encoder_decoder": true,
17
- "is_gated_act": false,
18
  "layer_norm_epsilon": 1e-06,
19
  "model_type": "t5",
20
  "n_positions": 512,
@@ -54,6 +54,7 @@
54
  "prefix": "translate English to Romanian: "
55
  }
56
  },
 
57
  "torch_dtype": "bfloat16",
58
  "transformers_version": "4.34.1",
59
  "use_cache": true,
 
1
  {
2
+ "_name_or_path": "google/flan-t5-base",
3
  "architectures": [
4
  "T5ForConditionalGeneration"
5
  ],
6
  "classifier_dropout": 0.0,
7
+ "d_ff": 2048,
8
  "d_kv": 64,
9
  "d_model": 768,
10
  "decoder_start_token_id": 0,
11
+ "dense_act_fn": "gelu_new",
12
  "dropout_rate": 0.1,
13
  "eos_token_id": 1,
14
+ "feed_forward_proj": "gated-gelu",
15
  "initializer_factor": 1.0,
16
  "is_encoder_decoder": true,
17
+ "is_gated_act": true,
18
  "layer_norm_epsilon": 1e-06,
19
  "model_type": "t5",
20
  "n_positions": 512,
 
54
  "prefix": "translate English to Romanian: "
55
  }
56
  },
57
+ "tie_word_embeddings": false,
58
  "torch_dtype": "bfloat16",
59
  "transformers_version": "4.34.1",
60
  "use_cache": true,
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c476c55da5595c38981cc465c90cb28010e06d3fe28406f1a997fa4322d4acc5
3
- size 445892686
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b82f8a503083930313171a7610e0ab4024cc7dc9f10ff08f75cfde897b0f11c8
3
+ size 495249650