Upload 9 files
Browse files- alpha.pt +3 -0
- config.json +33 -0
- dict.txt +0 -0
- openmatch_config.json +10 -0
- pytorch_model.bin +3 -0
- sp.model +3 -0
- special_tokens_map.json +8 -0
- tokenizer_config.json +13 -0
- training_args.bin +3 -0
alpha.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fbd70151b5430d6171734d22ac9f7b801127771eeff9eddeb19a15be90324514
|
3 |
+
size 81005
|
config.json
ADDED
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "/data/user_data/peixuanh/ckpts/ClueWeb22_All/DRO_emb_web_graph_url_RE",
|
3 |
+
"architectures": [
|
4 |
+
"T5Model"
|
5 |
+
],
|
6 |
+
"d_ff": 3072,
|
7 |
+
"d_kv": 64,
|
8 |
+
"d_model": 768,
|
9 |
+
"decoder_start_token_id": 0,
|
10 |
+
"dropout_rate": 0.1,
|
11 |
+
"eos_token_id": 2,
|
12 |
+
"feed_forward_proj": "relu",
|
13 |
+
"id2label": {
|
14 |
+
"0": "LABEL_0"
|
15 |
+
},
|
16 |
+
"initializer_factor": 1.0,
|
17 |
+
"is_encoder_decoder": true,
|
18 |
+
"label2id": {
|
19 |
+
"LABEL_0": 0
|
20 |
+
},
|
21 |
+
"layer_norm_epsilon": 1e-05,
|
22 |
+
"model_type": "t5",
|
23 |
+
"num_decoder_layers": 12,
|
24 |
+
"num_heads": 12,
|
25 |
+
"num_layers": 12,
|
26 |
+
"output_past": true,
|
27 |
+
"pad_token_id": 1,
|
28 |
+
"relative_attention_num_buckets": 32,
|
29 |
+
"torch_dtype": "float32",
|
30 |
+
"transformers_version": "4.21.0.dev0",
|
31 |
+
"use_cache": true,
|
32 |
+
"vocab_size": 33280
|
33 |
+
}
|
dict.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
openmatch_config.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"tied": true,
|
3 |
+
"plm_backbone": {
|
4 |
+
"type": "T5Model",
|
5 |
+
"feature": "last_hidden_state"
|
6 |
+
},
|
7 |
+
"pooling": "first",
|
8 |
+
"linear_head": false,
|
9 |
+
"normalize": false
|
10 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ffddda52332b8ec1f69c78638b1663d06251c41e3e51549b99b39d322930bc55
|
3 |
+
size 902652409
|
sp.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1f7dda6f68d6676a944418205def0e463eaa7b9aa3868f99b1ac122c849af152
|
3 |
+
size 824732
|
special_tokens_map.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<s>",
|
3 |
+
"cls_token": "<s>",
|
4 |
+
"eos_token": "</s>",
|
5 |
+
"pad_token": "<pad>",
|
6 |
+
"sep_token": "</s>",
|
7 |
+
"unk_token": "<unk>"
|
8 |
+
}
|
tokenizer_config.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<s>",
|
3 |
+
"cls_token": "<s>",
|
4 |
+
"eos_token": "</s>",
|
5 |
+
"lower": true,
|
6 |
+
"n_sentinel_tokens": 512,
|
7 |
+
"name_or_path": "/data/user_data/peixuanh/ckpts/ClueWeb22_All/DRO_emb_web_graph_url_RE",
|
8 |
+
"pad_token": "<pad>",
|
9 |
+
"sep_token": "</s>",
|
10 |
+
"special_tokens_map_file": "/data/user_data/peixuanh/models/web_continuous_t5_ckpt/special_tokens_map.json",
|
11 |
+
"tokenizer_class": "FairseqT5Tokenizer",
|
12 |
+
"unk_token": "<unk>"
|
13 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:dac0df07574549f36f689068ae6e77dbccee63c08d0fad93540e85cb14a5d50b
|
3 |
+
size 4091
|