Training in progress, epoch 1
Browse files- .gitignore +1 -0
- config.json +61 -0
- logs/1715834111.93729/events.out.tfevents.1715834111.cool101 +3 -0
- logs/1715834186.8552299/events.out.tfevents.1715834186.cool101 +3 -0
- logs/1715834584.4643378/events.out.tfevents.1715834584.cool101 +3 -0
- logs/1715834681.6287467/events.out.tfevents.1715834681.cool101 +3 -0
- logs/1715834885.759075/events.out.tfevents.1715834885.cool101 +3 -0
- logs/1715835317.7883132/events.out.tfevents.1715835317.cool101 +3 -0
- logs/1715835530.5656266/events.out.tfevents.1715835530.cool101 +3 -0
- logs/events.out.tfevents.1715834111.cool101 +3 -0
- logs/events.out.tfevents.1715834186.cool101 +3 -0
- logs/events.out.tfevents.1715834584.cool101 +3 -0
- logs/events.out.tfevents.1715834681.cool101 +3 -0
- logs/events.out.tfevents.1715834885.cool101 +3 -0
- logs/events.out.tfevents.1715835317.cool101 +3 -0
- logs/events.out.tfevents.1715835530.cool101 +3 -0
- pytorch_model.bin +3 -0
- training_args.bin +3 -0
.gitignore
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
checkpoint-*/
|
config.json
ADDED
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "google/flan-t5-base",
|
3 |
+
"architectures": [
|
4 |
+
"T5ForConditionalGeneration"
|
5 |
+
],
|
6 |
+
"d_ff": 2048,
|
7 |
+
"d_kv": 64,
|
8 |
+
"d_model": 768,
|
9 |
+
"decoder_start_token_id": 0,
|
10 |
+
"dense_act_fn": "gelu_new",
|
11 |
+
"dropout_rate": 0.1,
|
12 |
+
"eos_token_id": 1,
|
13 |
+
"feed_forward_proj": "gated-gelu",
|
14 |
+
"initializer_factor": 1.0,
|
15 |
+
"is_encoder_decoder": true,
|
16 |
+
"is_gated_act": true,
|
17 |
+
"layer_norm_epsilon": 1e-06,
|
18 |
+
"model_type": "t5",
|
19 |
+
"n_positions": 512,
|
20 |
+
"num_decoder_layers": 12,
|
21 |
+
"num_heads": 12,
|
22 |
+
"num_layers": 12,
|
23 |
+
"output_past": true,
|
24 |
+
"pad_token_id": 0,
|
25 |
+
"relative_attention_max_distance": 128,
|
26 |
+
"relative_attention_num_buckets": 32,
|
27 |
+
"task_specific_params": {
|
28 |
+
"summarization": {
|
29 |
+
"early_stopping": true,
|
30 |
+
"length_penalty": 2.0,
|
31 |
+
"max_length": 200,
|
32 |
+
"min_length": 30,
|
33 |
+
"no_repeat_ngram_size": 3,
|
34 |
+
"num_beams": 4,
|
35 |
+
"prefix": "summarize: "
|
36 |
+
},
|
37 |
+
"translation_en_to_de": {
|
38 |
+
"early_stopping": true,
|
39 |
+
"max_length": 300,
|
40 |
+
"num_beams": 4,
|
41 |
+
"prefix": "translate English to German: "
|
42 |
+
},
|
43 |
+
"translation_en_to_fr": {
|
44 |
+
"early_stopping": true,
|
45 |
+
"max_length": 300,
|
46 |
+
"num_beams": 4,
|
47 |
+
"prefix": "translate English to French: "
|
48 |
+
},
|
49 |
+
"translation_en_to_ro": {
|
50 |
+
"early_stopping": true,
|
51 |
+
"max_length": 300,
|
52 |
+
"num_beams": 4,
|
53 |
+
"prefix": "translate English to Romanian: "
|
54 |
+
}
|
55 |
+
},
|
56 |
+
"tie_word_embeddings": false,
|
57 |
+
"torch_dtype": "float32",
|
58 |
+
"transformers_version": "4.28.1",
|
59 |
+
"use_cache": true,
|
60 |
+
"vocab_size": 32128
|
61 |
+
}
|
logs/1715834111.93729/events.out.tfevents.1715834111.cool101
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a4e7bd0782727d665dad2af75d3d5a0e1cb8fea22d3fef27121cbe923194e422
|
3 |
+
size 5905
|
logs/1715834186.8552299/events.out.tfevents.1715834186.cool101
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:62790984d8715882a458c8cad4e8996cb3162538742dd7198c75dec1715ee159
|
3 |
+
size 5905
|
logs/1715834584.4643378/events.out.tfevents.1715834584.cool101
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9f2b6fa9fd5a4a90ce17d0b4b8761965746bbc54dd1e84d51fcb2def74099515
|
3 |
+
size 5905
|
logs/1715834681.6287467/events.out.tfevents.1715834681.cool101
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:890e8193f100f722ab7fbe0762f765361ba3be592a7c30448a04242a08cece46
|
3 |
+
size 5905
|
logs/1715834885.759075/events.out.tfevents.1715834885.cool101
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6a8ff5147d58c92ac6acae0672a83a6667ad471e5b0197abd7320d54f8b0fdf8
|
3 |
+
size 5905
|
logs/1715835317.7883132/events.out.tfevents.1715835317.cool101
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b30c91e3be047e193aa492752671abbd2ea8bda8f6805d0a559ff5bbdbf1e5b3
|
3 |
+
size 5905
|
logs/1715835530.5656266/events.out.tfevents.1715835530.cool101
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9b87413991acbf3874842033f4df57ad8062fd1719cdede1c72455f9e2ac0793
|
3 |
+
size 5905
|
logs/events.out.tfevents.1715834111.cool101
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:38f154a2963fd81c1ad33158cfa849d747f28432154076b29a0a3b2c7f3c6b52
|
3 |
+
size 5017
|
logs/events.out.tfevents.1715834186.cool101
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:24a60e463494163bd015c3820fd92832eb52108cc906a6fdaf8171bae2fd9d44
|
3 |
+
size 5017
|
logs/events.out.tfevents.1715834584.cool101
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2145fb814115ffdde59d19eabf9e274ad05c8d9134b5983cdc9f51bb7808e1fb
|
3 |
+
size 5017
|
logs/events.out.tfevents.1715834681.cool101
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3529fa2825acb95f508ed4110b629bf82231902579e4384fee9ae973cede7865
|
3 |
+
size 5017
|
logs/events.out.tfevents.1715834885.cool101
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:14562e7e65c43c94e831fec0b2dcc37a54c798c99eb62cbeed0db0abcb581212
|
3 |
+
size 5017
|
logs/events.out.tfevents.1715835317.cool101
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3063cda7a7559f2be118219ae889787f1854abd39ffc289e803fdf6127b43697
|
3 |
+
size 5017
|
logs/events.out.tfevents.1715835530.cool101
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4e82afb7731aeba2c044d55386a278ba8104da59bc92b6f96650ecf2f9115bac
|
3 |
+
size 5017
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4a21281724a236d425ab1a7cccbc8077e7dad0a54c2fce146c3b53db063ff1e2
|
3 |
+
size 990409330
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0dedc2dae0e5a224c4fe4c54ac88674c0b6cfc212e61aaf5a4e7dc045fbe90bf
|
3 |
+
size 4216
|