saketh092 commited on
Commit
ae76e77
1 Parent(s): 0abd41b

Upload T5ForConditionalGeneration

Browse files
Files changed (2) hide show
  1. config.json +9 -10
  2. pytorch_model.bin +2 -2
config.json CHANGED
@@ -1,25 +1,25 @@
1
  {
2
- "_name_or_path": "google/flan-t5-small",
3
  "architectures": [
4
  "T5ForConditionalGeneration"
5
  ],
6
- "d_ff": 1024,
7
  "d_kv": 64,
8
- "d_model": 512,
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
  "model_type": "t5",
19
  "n_positions": 512,
20
- "num_decoder_layers": 8,
21
- "num_heads": 6,
22
- "num_layers": 8,
23
  "output_past": true,
24
  "pad_token_id": 0,
25
  "relative_attention_max_distance": 128,
@@ -53,7 +53,6 @@
53
  "prefix": "translate English to Romanian: "
54
  }
55
  },
56
- "tie_word_embeddings": false,
57
  "torch_dtype": "float16",
58
  "transformers_version": "4.31.0",
59
  "use_cache": true,
 
1
  {
2
+ "_name_or_path": "google-t5/t5-base",
3
  "architectures": [
4
  "T5ForConditionalGeneration"
5
  ],
6
+ "d_ff": 3072,
7
  "d_kv": 64,
8
+ "d_model": 768,
9
  "decoder_start_token_id": 0,
10
+ "dense_act_fn": "relu",
11
  "dropout_rate": 0.1,
12
  "eos_token_id": 1,
13
+ "feed_forward_proj": "relu",
14
  "initializer_factor": 1.0,
15
  "is_encoder_decoder": true,
16
+ "is_gated_act": false,
17
  "layer_norm_epsilon": 1e-06,
18
  "model_type": "t5",
19
  "n_positions": 512,
20
+ "num_decoder_layers": 12,
21
+ "num_heads": 12,
22
+ "num_layers": 12,
23
  "output_past": true,
24
  "pad_token_id": 0,
25
  "relative_attention_max_distance": 128,
 
53
  "prefix": "translate English to Romanian: "
54
  }
55
  },
 
56
  "torch_dtype": "float16",
57
  "transformers_version": "4.31.0",
58
  "use_cache": true,
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d6a72b686a491d9f1dc4bc542fa5cdb4adffc49421cf22b10884874571bead66
3
- size 170765506
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fec8352e7a9bc0db85b7e74c97ef668c02b8dbbd102cd58896d98d37cb8823c5
3
+ size 559142542