Training in progress, step 500
Browse files- .gitattributes +1 -0
- .gitignore +1 -0
- config.json +29 -0
- pytorch_model.bin +3 -0
- runs/May17_14-16-21_92f3b98c3121/1652797198.8432558/events.out.tfevents.1652797198.92f3b98c3121.81.1 +3 -0
- runs/May17_14-16-21_92f3b98c3121/events.out.tfevents.1652797198.92f3b98c3121.81.0 +3 -0
- runs/May17_14-20-26_92f3b98c3121/1652797255.2656431/events.out.tfevents.1652797255.92f3b98c3121.81.3 +3 -0
- runs/May17_14-20-26_92f3b98c3121/events.out.tfevents.1652797255.92f3b98c3121.81.2 +3 -0
- runs/May17_14-25-05_92f3b98c3121/1652797530.1651053/events.out.tfevents.1652797530.92f3b98c3121.81.5 +3 -0
- runs/May17_14-25-05_92f3b98c3121/events.out.tfevents.1652797530.92f3b98c3121.81.4 +3 -0
- runs/May17_14-32-50_92f3b98c3121/1652798003.6830597/events.out.tfevents.1652798003.92f3b98c3121.81.7 +3 -0
- runs/May17_14-32-50_92f3b98c3121/events.out.tfevents.1652798003.92f3b98c3121.81.6 +3 -0
- special_tokens_map.json +1 -0
- spiece.model +3 -0
- tokenizer.json +3 -0
- tokenizer_config.json +1 -0
- training_args.bin +3 -0
.gitattributes
CHANGED
@@ -25,3 +25,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
25 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
26 |
*.zstandard filter=lfs diff=lfs merge=lfs -text
|
27 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
25 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
26 |
*.zstandard filter=lfs diff=lfs merge=lfs -text
|
27 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
28 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
.gitignore
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
checkpoint-*/
|
config.json
ADDED
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "google/mt5-small",
|
3 |
+
"architectures": [
|
4 |
+
"MT5ForConditionalGeneration"
|
5 |
+
],
|
6 |
+
"d_ff": 1024,
|
7 |
+
"d_kv": 64,
|
8 |
+
"d_model": 512,
|
9 |
+
"decoder_start_token_id": 0,
|
10 |
+
"dropout_rate": 0.1,
|
11 |
+
"eos_token_id": 1,
|
12 |
+
"feed_forward_proj": "gated-gelu",
|
13 |
+
"initializer_factor": 1.0,
|
14 |
+
"is_encoder_decoder": true,
|
15 |
+
"layer_norm_epsilon": 1e-06,
|
16 |
+
"model_type": "mt5",
|
17 |
+
"num_decoder_layers": 8,
|
18 |
+
"num_heads": 6,
|
19 |
+
"num_layers": 8,
|
20 |
+
"pad_token_id": 0,
|
21 |
+
"relative_attention_max_distance": 128,
|
22 |
+
"relative_attention_num_buckets": 32,
|
23 |
+
"tie_word_embeddings": false,
|
24 |
+
"tokenizer_class": "T5Tokenizer",
|
25 |
+
"torch_dtype": "float32",
|
26 |
+
"transformers_version": "4.19.2",
|
27 |
+
"use_cache": true,
|
28 |
+
"vocab_size": 250112
|
29 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:daaa899f6b54b68dab5041382469e05a98af0bd071586828ba57f91662e6bd1c
|
3 |
+
size 1200770885
|
runs/May17_14-16-21_92f3b98c3121/1652797198.8432558/events.out.tfevents.1652797198.92f3b98c3121.81.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b28ea3c532332f513a6d257209e494087d7229058b1756ef59be053b302265fd
|
3 |
+
size 5470
|
runs/May17_14-16-21_92f3b98c3121/events.out.tfevents.1652797198.92f3b98c3121.81.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:52fddf6040abc43157dd9d96d4d7f56bb2ba9fd845e6370cad50a737f31cfeec
|
3 |
+
size 3759
|
runs/May17_14-20-26_92f3b98c3121/1652797255.2656431/events.out.tfevents.1652797255.92f3b98c3121.81.3
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:54c966386b3ff3ce30c5db7b71bdc8a0d131a0610417962ce45f294494e8dbd8
|
3 |
+
size 5470
|
runs/May17_14-20-26_92f3b98c3121/events.out.tfevents.1652797255.92f3b98c3121.81.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:36386a197ebad6ac6b8bd0df2718404c324f157a9ef7bb3e83770314b4a5c811
|
3 |
+
size 3759
|
runs/May17_14-25-05_92f3b98c3121/1652797530.1651053/events.out.tfevents.1652797530.92f3b98c3121.81.5
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2a4aaba919e4d564cc0bddbf8f1d759c6a9d479ea53bf2ff7ce8ce51a87f7544
|
3 |
+
size 5470
|
runs/May17_14-25-05_92f3b98c3121/events.out.tfevents.1652797530.92f3b98c3121.81.4
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:de5918439f9a624029db6f96bbb57e64f1ee7541513920b6f517ed81ab1f6453
|
3 |
+
size 3758
|
runs/May17_14-32-50_92f3b98c3121/1652798003.6830597/events.out.tfevents.1652798003.92f3b98c3121.81.7
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f1987ccaa8213606a54937a91a5c63c3300d2fed4b8ba9a50487a0db87f8ff30
|
3 |
+
size 5470
|
runs/May17_14-32-50_92f3b98c3121/events.out.tfevents.1652798003.92f3b98c3121.81.6
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:21ecdafd3daef8a1b8189b4f847393d3cfe1e7e10e17af347417af4b1606d3ec
|
3 |
+
size 11572
|
special_tokens_map.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}
|
spiece.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ef78f86560d809067d12bac6c09f19a462cb3af3f54d2b8acbba26e1433125d6
|
3 |
+
size 4309802
|
tokenizer.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:faaa6405f5f79c9e788c7980874a9a3b5b0aea07b53bd9243bf1abb8f5c49c81
|
3 |
+
size 16330467
|
tokenizer_config.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "extra_ids": 0, "additional_special_tokens": null, "special_tokens_map_file": "/root/.cache/huggingface/transformers/685ac0ca8568ec593a48b61b0a3c272beee9bc194a3c7241d15dcadb5f875e53.f76030f3ec1b96a8199b2593390c610e76ca8028ef3d24680000619ffb646276", "name_or_path": "google/mt5-small", "sp_model_kwargs": {}, "tokenizer_class": "T5Tokenizer"}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5be4ee5b9704babe33821ae127a6f65a7e2a476027ea5cb3321309624eec632c
|
3 |
+
size 3375
|