dokyoungkim
commited on
Commit
•
315759c
1
Parent(s):
a2fcd97
Training in progress, epoch 1
Browse files- config.json +48 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- runs/Mar18_19-30-00_gpusvr021320/events.out.tfevents.1710757951.gpusvr021320.753078.0 +3 -0
- runs/Mar18_19-44-31_gpusvr021320/events.out.tfevents.1710758843.gpusvr021320.756466.0 +3 -0
- runs/Mar18_20-08-17_gpusvr021320/events.out.tfevents.1710760258.gpusvr021320.765054.0 +3 -0
- runs/Mar18_20-11-50_gpusvr021320/events.out.tfevents.1710760369.gpusvr021320.766285.0 +3 -0
- runs/Mar18_20-13-16_gpusvr021320/events.out.tfevents.1710760539.gpusvr021320.767049.0 +3 -0
- runs/Mar18_20-30-53_gpusvr021320/events.out.tfevents.1710761638.gpusvr021320.773013.0 +3 -0
- runs/Mar18_20-40-40_gpusvr021320/events.out.tfevents.1710762051.gpusvr021320.773013.1 +3 -0
- runs/Mar18_20-42-29_gpusvr021320/events.out.tfevents.1710762152.gpusvr021320.783960.0 +3 -0
- runs/Mar18_20-45-16_gpusvr021320/events.out.tfevents.1710762319.gpusvr021320.801134.0 +3 -0
- special_tokens_map.json +6 -0
- tokenizer_config.json +48 -0
- training_args.bin +3 -0
- vocab-src.json +0 -0
- vocab-tgt.json +0 -0
config.json
ADDED
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "facebook/wmt19-de-en",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"activation_function": "relu",
|
5 |
+
"architectures": [
|
6 |
+
"FSMTForConditionalGeneration"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.1,
|
9 |
+
"bos_token_id": 0,
|
10 |
+
"d_model": 1024,
|
11 |
+
"decoder": {
|
12 |
+
"bos_token_id": 2,
|
13 |
+
"model_type": "fsmt_decoder",
|
14 |
+
"vocab_size": 42024
|
15 |
+
},
|
16 |
+
"decoder_attention_heads": 16,
|
17 |
+
"decoder_ffn_dim": 4096,
|
18 |
+
"decoder_layerdrop": 0,
|
19 |
+
"decoder_layers": 6,
|
20 |
+
"decoder_start_token_id": 2,
|
21 |
+
"dropout": 0.2,
|
22 |
+
"encoder_attention_heads": 16,
|
23 |
+
"encoder_ffn_dim": 8192,
|
24 |
+
"encoder_layerdrop": 0,
|
25 |
+
"encoder_layers": 6,
|
26 |
+
"eos_token_id": 2,
|
27 |
+
"forced_eos_token_id": 2,
|
28 |
+
"init_std": 0.02,
|
29 |
+
"is_encoder_decoder": true,
|
30 |
+
"langs": [
|
31 |
+
"de",
|
32 |
+
"en"
|
33 |
+
],
|
34 |
+
"length_penalty": 1.1,
|
35 |
+
"max_length": 200,
|
36 |
+
"max_position_embeddings": 1024,
|
37 |
+
"model_type": "fsmt",
|
38 |
+
"num_beams": 5,
|
39 |
+
"num_hidden_layers": 6,
|
40 |
+
"pad_token_id": 1,
|
41 |
+
"scale_embedding": true,
|
42 |
+
"src_vocab_size": 42024,
|
43 |
+
"tgt_vocab_size": 42024,
|
44 |
+
"tie_word_embeddings": true,
|
45 |
+
"torch_dtype": "float32",
|
46 |
+
"transformers_version": "4.37.2",
|
47 |
+
"use_cache": true
|
48 |
+
}
|
merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f759d278376d01632df4cae53ee1536ef2ef90859752f1cc0a891b939ea52c70
|
3 |
+
size 1087419504
|
runs/Mar18_19-30-00_gpusvr021320/events.out.tfevents.1710757951.gpusvr021320.753078.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e6d23736c4d8662101f7d7f36fe3ef724429e6022b9492fdb72c5ba2ead487da
|
3 |
+
size 342
|
runs/Mar18_19-44-31_gpusvr021320/events.out.tfevents.1710758843.gpusvr021320.756466.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9ef96e9ec4c2a3a87f9f699b9d26d6d58250a965920d6e4b41dcecdefd54c8de
|
3 |
+
size 342
|
runs/Mar18_20-08-17_gpusvr021320/events.out.tfevents.1710760258.gpusvr021320.765054.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:39ffcff5b51f5dfc9f7da8eb14998513bb32f4c5534a695f4644e4cda95d61db
|
3 |
+
size 342
|
runs/Mar18_20-11-50_gpusvr021320/events.out.tfevents.1710760369.gpusvr021320.766285.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4308f4ca63e5a2637a9bce29b612c43a783e07c7fb0d5cb9e6a960a33d7e0385
|
3 |
+
size 342
|
runs/Mar18_20-13-16_gpusvr021320/events.out.tfevents.1710760539.gpusvr021320.767049.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cd34cfbf253d6a0f153306ffc552ccc8d28d65d1e89bba6c4193c84b8a0559e7
|
3 |
+
size 342
|
runs/Mar18_20-30-53_gpusvr021320/events.out.tfevents.1710761638.gpusvr021320.773013.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:347d2ae304796e66304c0be0630158025e43fb889359a1a1d246a4cfc009081e
|
3 |
+
size 342
|
runs/Mar18_20-40-40_gpusvr021320/events.out.tfevents.1710762051.gpusvr021320.773013.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:389a085d25fd48844cf8327cc84d1cd7aaf116dffe2866bd3aa5716335693053
|
3 |
+
size 4184
|
runs/Mar18_20-42-29_gpusvr021320/events.out.tfevents.1710762152.gpusvr021320.783960.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9bc21b32e632e9709e1031b2dcc81a64d5b4fa0a375dcd5f673dfb9ea05c1528
|
3 |
+
size 4184
|
runs/Mar18_20-45-16_gpusvr021320/events.out.tfevents.1710762319.gpusvr021320.801134.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:09731ec86686ef03cd5d3bdbe30491315819cc8f2dcbc02271e01f69e4b77c02
|
3 |
+
size 6732
|
special_tokens_map.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<s>",
|
3 |
+
"pad_token": "<pad>",
|
4 |
+
"sep_token": "</s>",
|
5 |
+
"unk_token": "<unk>"
|
6 |
+
}
|
tokenizer_config.json
ADDED
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "<s>",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"1": {
|
12 |
+
"content": "<pad>",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"2": {
|
20 |
+
"content": "</s>",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"3": {
|
28 |
+
"content": "<unk>",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
}
|
35 |
+
},
|
36 |
+
"bos_token": "<s>",
|
37 |
+
"clean_up_tokenization_spaces": true,
|
38 |
+
"do_lower_case": false,
|
39 |
+
"langs": [
|
40 |
+
"de",
|
41 |
+
"en"
|
42 |
+
],
|
43 |
+
"model_max_length": 1024,
|
44 |
+
"pad_token": "<pad>",
|
45 |
+
"sep_token": "</s>",
|
46 |
+
"tokenizer_class": "FSMTTokenizer",
|
47 |
+
"unk_token": "<unk>"
|
48 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c1de7d234c264ccb30ba3dc1e4be1bdfe33901ec174695a8a61d0b2798bbc7ff
|
3 |
+
size 4475
|
vocab-src.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
vocab-tgt.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|