Upload T5ForConditionalGeneration
Browse files- README.md +1 -1
- config.json +34 -0
- generation_config.json +7 -0
- model.safetensors +3 -0
README.md
CHANGED
|
@@ -1,5 +1,4 @@
|
|
| 1 |
---
|
| 2 |
-
license: mit
|
| 3 |
language:
|
| 4 |
- af
|
| 5 |
- am
|
|
@@ -104,6 +103,7 @@ language:
|
|
| 104 |
- yo
|
| 105 |
- zh
|
| 106 |
- zu
|
|
|
|
| 107 |
datasets:
|
| 108 |
- mc4
|
| 109 |
---
|
|
|
|
| 1 |
---
|
|
|
|
| 2 |
language:
|
| 3 |
- af
|
| 4 |
- am
|
|
|
|
| 103 |
- yo
|
| 104 |
- zh
|
| 105 |
- zu
|
| 106 |
+
license: mit
|
| 107 |
datasets:
|
| 108 |
- mc4
|
| 109 |
---
|
config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "../../hf_checkpoints/myt5_base_250000",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"T5ForConditionalGeneration"
|
| 5 |
+
],
|
| 6 |
+
"classifier_dropout": 0.0,
|
| 7 |
+
"d_ff": 3968,
|
| 8 |
+
"d_kv": 64,
|
| 9 |
+
"d_model": 1536,
|
| 10 |
+
"decoder_start_token_id": 0,
|
| 11 |
+
"dense_act_fn": "gelu_new",
|
| 12 |
+
"dropout_rate": 0.0,
|
| 13 |
+
"eos_token_id": 1,
|
| 14 |
+
"feed_forward_proj": "gated-gelu",
|
| 15 |
+
"gradient_checkpointing": false,
|
| 16 |
+
"initializer_factor": 1.0,
|
| 17 |
+
"is_encoder_decoder": true,
|
| 18 |
+
"is_gated_act": true,
|
| 19 |
+
"layer_norm_epsilon": 1e-06,
|
| 20 |
+
"model_type": "t5",
|
| 21 |
+
"num_decoder_layers": 6,
|
| 22 |
+
"num_heads": 12,
|
| 23 |
+
"num_layers": 18,
|
| 24 |
+
"output_past": true,
|
| 25 |
+
"pad_token_id": 0,
|
| 26 |
+
"relative_attention_max_distance": 128,
|
| 27 |
+
"relative_attention_num_buckets": 32,
|
| 28 |
+
"tie_word_embeddings": false,
|
| 29 |
+
"tokenizer_class": "ByT5Tokenizer",
|
| 30 |
+
"torch_dtype": "float32",
|
| 31 |
+
"transformers_version": "4.38.1",
|
| 32 |
+
"use_cache": true,
|
| 33 |
+
"vocab_size": 384
|
| 34 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"decoder_start_token_id": 0,
|
| 4 |
+
"eos_token_id": 1,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.38.1"
|
| 7 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5e0614f25c67dfa0a9045f77cf96e90625e7cfeb49c5df7971e3ae792bcd3ddf
|
| 3 |
+
size 2326643632
|