Upload T5ForConditionalGeneration
Browse files- config.json +53 -0
- generation_config.json +7 -0
- pytorch_model.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "analogy_models/opt-350m-analogy-epoch15/model",
|
| 3 |
+
"_remove_final_layer_norm": false,
|
| 4 |
+
"activation_dropout": 0.0,
|
| 5 |
+
"activation_function": "relu",
|
| 6 |
+
"architectures": [
|
| 7 |
+
"T5ForConditionalGeneration"
|
| 8 |
+
],
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"bos_token_id": 2,
|
| 11 |
+
"d_ff": 2048,
|
| 12 |
+
"d_kv": 64,
|
| 13 |
+
"d_model": 1024,
|
| 14 |
+
"dense_act_fn": "relu",
|
| 15 |
+
"do_layer_norm_before": false,
|
| 16 |
+
"dropout": 0.1,
|
| 17 |
+
"dropout_rate": 0.1,
|
| 18 |
+
"enable_bias": true,
|
| 19 |
+
"eos_token_id": 2,
|
| 20 |
+
"feed_forward_proj": "relu",
|
| 21 |
+
"ffn_dim": 4096,
|
| 22 |
+
"finetuing_config": {
|
| 23 |
+
"batch_size": 32,
|
| 24 |
+
"data": "relbert/semeval2012_relational_similarity",
|
| 25 |
+
"epoch": 15,
|
| 26 |
+
"gradient_accumulation_steps": 1,
|
| 27 |
+
"learning_rate": 0.0001,
|
| 28 |
+
"model": "facebook/opt-350m",
|
| 29 |
+
"random_seed": 42,
|
| 30 |
+
"template": "<subj-a> is to <obj-a> what <subj-b> is to <obj-b>"
|
| 31 |
+
},
|
| 32 |
+
"init_std": 0.02,
|
| 33 |
+
"initializer_factor": 1.0,
|
| 34 |
+
"is_encoder_decoder": true,
|
| 35 |
+
"is_gated_act": false,
|
| 36 |
+
"layer_norm_elementwise_affine": true,
|
| 37 |
+
"layer_norm_epsilon": 1e-06,
|
| 38 |
+
"layerdrop": 0.0,
|
| 39 |
+
"max_position_embeddings": 2048,
|
| 40 |
+
"model_type": "t5",
|
| 41 |
+
"num_decoder_layers": 6,
|
| 42 |
+
"num_heads": 16,
|
| 43 |
+
"num_layers": 24,
|
| 44 |
+
"pad_token_id": 1,
|
| 45 |
+
"prefix": "</s>",
|
| 46 |
+
"relative_attention_max_distance": 128,
|
| 47 |
+
"relative_attention_num_buckets": 32,
|
| 48 |
+
"torch_dtype": "float32",
|
| 49 |
+
"transformers_version": "4.26.1",
|
| 50 |
+
"use_cache": true,
|
| 51 |
+
"vocab_size": 50272,
|
| 52 |
+
"word_embed_proj_dim": 512
|
| 53 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 2,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"pad_token_id": 1,
|
| 6 |
+
"transformers_version": "4.26.1"
|
| 7 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a623d8eb8a964ad39669782686f2a0d3cdbd6a9319e06c5a06d1beb0fb0436fa
|
| 3 |
+
size 1313584089
|