Upload folder using huggingface_hub
Browse files- .gitattributes +6 -0
- README.md +35 -0
- checkpoint-24328/config.json +56 -0
- checkpoint-24328/generation_config.json +16 -0
- checkpoint-24328/model.safetensors +3 -0
- checkpoint-24328/optimizer.pt +3 -0
- checkpoint-24328/rng_state_0.pth +3 -0
- checkpoint-24328/rng_state_1.pth +3 -0
- checkpoint-24328/scheduler.pt +3 -0
- checkpoint-24328/source.spm +3 -0
- checkpoint-24328/special_tokens_map.json +5 -0
- checkpoint-24328/target.spm +3 -0
- checkpoint-24328/tokenizer_config.json +39 -0
- checkpoint-24328/trainer_state.json +404 -0
- checkpoint-24328/training_args.bin +3 -0
- checkpoint-24328/vocab.json +0 -0
- checkpoint-36492/config.json +56 -0
- checkpoint-36492/generation_config.json +16 -0
- checkpoint-36492/model.safetensors +3 -0
- checkpoint-36492/optimizer.pt +3 -0
- checkpoint-36492/rng_state_0.pth +3 -0
- checkpoint-36492/rng_state_1.pth +3 -0
- checkpoint-36492/scheduler.pt +3 -0
- checkpoint-36492/source.spm +3 -0
- checkpoint-36492/special_tokens_map.json +5 -0
- checkpoint-36492/target.spm +3 -0
- checkpoint-36492/tokenizer_config.json +39 -0
- checkpoint-36492/trainer_state.json +585 -0
- checkpoint-36492/training_args.bin +3 -0
- checkpoint-36492/vocab.json +0 -0
- config.json +56 -0
- generation_config.json +16 -0
- model.safetensors +3 -0
- source.spm +3 -0
- special_tokens_map.json +5 -0
- target.spm +3 -0
- tokenizer_config.json +39 -0
- training_args.bin +3 -0
- training_params.json +37 -0
- vocab.json +0 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,9 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
checkpoint-24328/source.spm filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
checkpoint-24328/target.spm filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
checkpoint-36492/source.spm filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
checkpoint-36492/target.spm filter=lfs diff=lfs merge=lfs -text
|
| 40 |
+
source.spm filter=lfs diff=lfs merge=lfs -text
|
| 41 |
+
target.spm filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
---
|
| 3 |
+
library_name: transformers
|
| 4 |
+
tags:
|
| 5 |
+
- autotrain
|
| 6 |
+
- text2text-generation
|
| 7 |
+
base_model: Helsinki-NLP/opus-mt-en-ru
|
| 8 |
+
widget:
|
| 9 |
+
- text: "I love AutoTrain"
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Model Trained Using AutoTrain
|
| 13 |
+
|
| 14 |
+
- Problem type: Seq2Seq
|
| 15 |
+
|
| 16 |
+
## Validation Metrics
|
| 17 |
+
loss: 0.5498164296150208
|
| 18 |
+
|
| 19 |
+
rouge1: 32.6453
|
| 20 |
+
|
| 21 |
+
rouge2: 28.31
|
| 22 |
+
|
| 23 |
+
rougeL: 32.1048
|
| 24 |
+
|
| 25 |
+
rougeLsum: 32.1708
|
| 26 |
+
|
| 27 |
+
gen_len: 33.8306
|
| 28 |
+
|
| 29 |
+
runtime: 1530.1218
|
| 30 |
+
|
| 31 |
+
samples_per_second: 14.132
|
| 32 |
+
|
| 33 |
+
steps_per_second: 0.442
|
| 34 |
+
|
| 35 |
+
: 3.0
|
checkpoint-24328/config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Helsinki-NLP/opus-mt-en-ru",
|
| 3 |
+
"_num_labels": 3,
|
| 4 |
+
"activation_dropout": 0.0,
|
| 5 |
+
"activation_function": "swish",
|
| 6 |
+
"add_bias_logits": false,
|
| 7 |
+
"add_final_layer_norm": false,
|
| 8 |
+
"architectures": [
|
| 9 |
+
"MarianMTModel"
|
| 10 |
+
],
|
| 11 |
+
"attention_dropout": 0.0,
|
| 12 |
+
"bos_token_id": 0,
|
| 13 |
+
"classif_dropout": 0.0,
|
| 14 |
+
"classifier_dropout": 0.0,
|
| 15 |
+
"d_model": 512,
|
| 16 |
+
"decoder_attention_heads": 8,
|
| 17 |
+
"decoder_ffn_dim": 2048,
|
| 18 |
+
"decoder_layerdrop": 0.0,
|
| 19 |
+
"decoder_layers": 6,
|
| 20 |
+
"decoder_start_token_id": 62517,
|
| 21 |
+
"decoder_vocab_size": 62518,
|
| 22 |
+
"dropout": 0.1,
|
| 23 |
+
"encoder_attention_heads": 8,
|
| 24 |
+
"encoder_ffn_dim": 2048,
|
| 25 |
+
"encoder_layerdrop": 0.0,
|
| 26 |
+
"encoder_layers": 6,
|
| 27 |
+
"eos_token_id": 0,
|
| 28 |
+
"forced_eos_token_id": 0,
|
| 29 |
+
"id2label": {
|
| 30 |
+
"0": "LABEL_0",
|
| 31 |
+
"1": "LABEL_1",
|
| 32 |
+
"2": "LABEL_2"
|
| 33 |
+
},
|
| 34 |
+
"init_std": 0.02,
|
| 35 |
+
"is_encoder_decoder": true,
|
| 36 |
+
"label2id": {
|
| 37 |
+
"LABEL_0": 0,
|
| 38 |
+
"LABEL_1": 1,
|
| 39 |
+
"LABEL_2": 2
|
| 40 |
+
},
|
| 41 |
+
"max_length": null,
|
| 42 |
+
"max_position_embeddings": 512,
|
| 43 |
+
"model_type": "marian",
|
| 44 |
+
"normalize_before": false,
|
| 45 |
+
"normalize_embedding": false,
|
| 46 |
+
"num_beams": null,
|
| 47 |
+
"num_hidden_layers": 6,
|
| 48 |
+
"pad_token_id": 62517,
|
| 49 |
+
"scale_embedding": true,
|
| 50 |
+
"share_encoder_decoder_embeddings": true,
|
| 51 |
+
"static_position_embeddings": true,
|
| 52 |
+
"torch_dtype": "float32",
|
| 53 |
+
"transformers_version": "4.48.0",
|
| 54 |
+
"use_cache": false,
|
| 55 |
+
"vocab_size": 62518
|
| 56 |
+
}
|
checkpoint-24328/generation_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bad_words_ids": [
|
| 3 |
+
[
|
| 4 |
+
62517
|
| 5 |
+
]
|
| 6 |
+
],
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"decoder_start_token_id": 62517,
|
| 9 |
+
"eos_token_id": 0,
|
| 10 |
+
"forced_eos_token_id": 0,
|
| 11 |
+
"max_length": 512,
|
| 12 |
+
"num_beams": 4,
|
| 13 |
+
"pad_token_id": 62517,
|
| 14 |
+
"renormalize_logits": true,
|
| 15 |
+
"transformers_version": "4.48.0"
|
| 16 |
+
}
|
checkpoint-24328/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f6f5eef1d23257a753025aed5e386686bd8b7ab5dff373ae663be0b3be07bc92
|
| 3 |
+
size 304869976
|
checkpoint-24328/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:038a1951d0f8e7666720a62721a9af9b3297936481db72ce9cb8b0f92b47e378
|
| 3 |
+
size 609393274
|
checkpoint-24328/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:494c89fc9a0a31848381908a1476f97062c3b7ac43f5d026334eb3667939ea1e
|
| 3 |
+
size 14512
|
checkpoint-24328/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fa8b05e18c01f9892cc1a0dccf8d4ad9ec04f7a1ad47e787bab7483f83f5d698
|
| 3 |
+
size 14512
|
checkpoint-24328/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7444e21e803fb824eee408a2cde84cd09e218a6ef226645fd65f99a9d4fa88cb
|
| 3 |
+
size 1064
|
checkpoint-24328/source.spm
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:16bebef1389a0b8ab452772c4e35b9e605e5713f8ac7baa71ca701394eaa086d
|
| 3 |
+
size 802781
|
checkpoint-24328/special_tokens_map.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eos_token": "</s>",
|
| 3 |
+
"pad_token": "<pad>",
|
| 4 |
+
"unk_token": "<unk>"
|
| 5 |
+
}
|
checkpoint-24328/target.spm
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:745998e51ba5b058e38b7ac7765c25c43ed5c1c39cc92b27163b9b2e323c9d7c
|
| 3 |
+
size 1080169
|
checkpoint-24328/tokenizer_config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "<unk>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"62517": {
|
| 20 |
+
"content": "<pad>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
}
|
| 27 |
+
},
|
| 28 |
+
"clean_up_tokenization_spaces": false,
|
| 29 |
+
"eos_token": "</s>",
|
| 30 |
+
"extra_special_tokens": {},
|
| 31 |
+
"model_max_length": 512,
|
| 32 |
+
"pad_token": "<pad>",
|
| 33 |
+
"separate_vocabs": false,
|
| 34 |
+
"source_lang": "en",
|
| 35 |
+
"sp_model_kwargs": {},
|
| 36 |
+
"target_lang": "ru",
|
| 37 |
+
"tokenizer_class": "MarianTokenizer",
|
| 38 |
+
"unk_token": "<unk>"
|
| 39 |
+
}
|
checkpoint-24328/trainer_state.json
ADDED
|
@@ -0,0 +1,404 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.5606569051742554,
|
| 3 |
+
"best_model_checkpoint": "ck3-localization/checkpoint-24328",
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 24328,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.041104899704044726,
|
| 13 |
+
"grad_norm": 6.12867546081543,
|
| 14 |
+
"learning_rate": 2.7397260273972604e-06,
|
| 15 |
+
"loss": 2.3513,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.08220979940808945,
|
| 20 |
+
"grad_norm": 6.480109691619873,
|
| 21 |
+
"learning_rate": 5.479452054794521e-06,
|
| 22 |
+
"loss": 1.4689,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.12331469911213416,
|
| 27 |
+
"grad_norm": 4.740694046020508,
|
| 28 |
+
"learning_rate": 8.219178082191782e-06,
|
| 29 |
+
"loss": 1.19,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.1644195988161789,
|
| 34 |
+
"grad_norm": 5.501825332641602,
|
| 35 |
+
"learning_rate": 1.0958904109589042e-05,
|
| 36 |
+
"loss": 1.0318,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.20552449852022361,
|
| 41 |
+
"grad_norm": 5.322640419006348,
|
| 42 |
+
"learning_rate": 1.3698630136986302e-05,
|
| 43 |
+
"loss": 0.9556,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.24662939822426833,
|
| 48 |
+
"grad_norm": 5.773024559020996,
|
| 49 |
+
"learning_rate": 1.6438356164383563e-05,
|
| 50 |
+
"loss": 0.9137,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.28773429792831307,
|
| 55 |
+
"grad_norm": 4.7130656242370605,
|
| 56 |
+
"learning_rate": 1.9178082191780822e-05,
|
| 57 |
+
"loss": 0.8705,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.3288391976323578,
|
| 62 |
+
"grad_norm": 3.756901741027832,
|
| 63 |
+
"learning_rate": 1.9786858291212475e-05,
|
| 64 |
+
"loss": 0.8531,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.3699440973364025,
|
| 69 |
+
"grad_norm": 4.033214092254639,
|
| 70 |
+
"learning_rate": 1.948237013580172e-05,
|
| 71 |
+
"loss": 0.8138,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.41104899704044723,
|
| 76 |
+
"grad_norm": 4.564835071563721,
|
| 77 |
+
"learning_rate": 1.9177881980390966e-05,
|
| 78 |
+
"loss": 0.7946,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.45215389674449197,
|
| 83 |
+
"grad_norm": 4.850255012512207,
|
| 84 |
+
"learning_rate": 1.887339382498021e-05,
|
| 85 |
+
"loss": 0.7833,
|
| 86 |
+
"step": 5500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.49325879644853665,
|
| 90 |
+
"grad_norm": 4.716800689697266,
|
| 91 |
+
"learning_rate": 1.8568905669569456e-05,
|
| 92 |
+
"loss": 0.7662,
|
| 93 |
+
"step": 6000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.5343636961525814,
|
| 97 |
+
"grad_norm": 4.0613274574279785,
|
| 98 |
+
"learning_rate": 1.82644175141587e-05,
|
| 99 |
+
"loss": 0.7668,
|
| 100 |
+
"step": 6500
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.5754685958566261,
|
| 104 |
+
"grad_norm": 4.6606831550598145,
|
| 105 |
+
"learning_rate": 1.7959929358747943e-05,
|
| 106 |
+
"loss": 0.7518,
|
| 107 |
+
"step": 7000
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.6165734955606709,
|
| 111 |
+
"grad_norm": 4.745144844055176,
|
| 112 |
+
"learning_rate": 1.765544120333719e-05,
|
| 113 |
+
"loss": 0.7477,
|
| 114 |
+
"step": 7500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.6576783952647156,
|
| 118 |
+
"grad_norm": 4.414773464202881,
|
| 119 |
+
"learning_rate": 1.7350953047926437e-05,
|
| 120 |
+
"loss": 0.751,
|
| 121 |
+
"step": 8000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.6987832949687602,
|
| 125 |
+
"grad_norm": 4.531556606292725,
|
| 126 |
+
"learning_rate": 1.704646489251568e-05,
|
| 127 |
+
"loss": 0.7424,
|
| 128 |
+
"step": 8500
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.739888194672805,
|
| 132 |
+
"grad_norm": 4.841054439544678,
|
| 133 |
+
"learning_rate": 1.6741976737104928e-05,
|
| 134 |
+
"loss": 0.7328,
|
| 135 |
+
"step": 9000
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.7809930943768497,
|
| 139 |
+
"grad_norm": 4.302633762359619,
|
| 140 |
+
"learning_rate": 1.6437488581694175e-05,
|
| 141 |
+
"loss": 0.725,
|
| 142 |
+
"step": 9500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.8220979940808945,
|
| 146 |
+
"grad_norm": 3.530836343765259,
|
| 147 |
+
"learning_rate": 1.6133000426283418e-05,
|
| 148 |
+
"loss": 0.7072,
|
| 149 |
+
"step": 10000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.8632028937849392,
|
| 153 |
+
"grad_norm": 4.2016191482543945,
|
| 154 |
+
"learning_rate": 1.5828512270872665e-05,
|
| 155 |
+
"loss": 0.6895,
|
| 156 |
+
"step": 10500
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.9043077934889839,
|
| 160 |
+
"grad_norm": 3.463226079940796,
|
| 161 |
+
"learning_rate": 1.5524024115461912e-05,
|
| 162 |
+
"loss": 0.7093,
|
| 163 |
+
"step": 11000
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.9454126931930286,
|
| 167 |
+
"grad_norm": 3.3261220455169678,
|
| 168 |
+
"learning_rate": 1.5219535960051155e-05,
|
| 169 |
+
"loss": 0.6941,
|
| 170 |
+
"step": 11500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.9865175928970733,
|
| 174 |
+
"grad_norm": 3.778331756591797,
|
| 175 |
+
"learning_rate": 1.49150478046404e-05,
|
| 176 |
+
"loss": 0.7008,
|
| 177 |
+
"step": 12000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.0,
|
| 181 |
+
"eval_gen_len": 33.8514,
|
| 182 |
+
"eval_loss": 0.6079365611076355,
|
| 183 |
+
"eval_rouge1": 31.9733,
|
| 184 |
+
"eval_rouge2": 27.2686,
|
| 185 |
+
"eval_rougeL": 31.3526,
|
| 186 |
+
"eval_rougeLsum": 31.4543,
|
| 187 |
+
"eval_runtime": 1591.2259,
|
| 188 |
+
"eval_samples_per_second": 13.59,
|
| 189 |
+
"eval_steps_per_second": 0.425,
|
| 190 |
+
"step": 12164
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 1.0276224926011182,
|
| 194 |
+
"grad_norm": 3.7499120235443115,
|
| 195 |
+
"learning_rate": 1.4610559649229646e-05,
|
| 196 |
+
"loss": 0.6284,
|
| 197 |
+
"step": 12500
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 1.0687273923051628,
|
| 201 |
+
"grad_norm": 3.52150821685791,
|
| 202 |
+
"learning_rate": 1.4306071493818891e-05,
|
| 203 |
+
"loss": 0.6337,
|
| 204 |
+
"step": 13000
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 1.1098322920092074,
|
| 208 |
+
"grad_norm": 4.520621299743652,
|
| 209 |
+
"learning_rate": 1.4001583338408138e-05,
|
| 210 |
+
"loss": 0.6292,
|
| 211 |
+
"step": 13500
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 1.1509371917132523,
|
| 215 |
+
"grad_norm": 4.275709629058838,
|
| 216 |
+
"learning_rate": 1.3697095182997382e-05,
|
| 217 |
+
"loss": 0.6195,
|
| 218 |
+
"step": 14000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 1.192042091417297,
|
| 222 |
+
"grad_norm": 3.500743865966797,
|
| 223 |
+
"learning_rate": 1.3392607027586628e-05,
|
| 224 |
+
"loss": 0.6168,
|
| 225 |
+
"step": 14500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 1.2331469911213417,
|
| 229 |
+
"grad_norm": 2.587315320968628,
|
| 230 |
+
"learning_rate": 1.3088118872175875e-05,
|
| 231 |
+
"loss": 0.6169,
|
| 232 |
+
"step": 15000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 1.2742518908253864,
|
| 236 |
+
"grad_norm": 3.4539663791656494,
|
| 237 |
+
"learning_rate": 1.2783630716765119e-05,
|
| 238 |
+
"loss": 0.619,
|
| 239 |
+
"step": 15500
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 1.3153567905294312,
|
| 243 |
+
"grad_norm": 3.759758710861206,
|
| 244 |
+
"learning_rate": 1.2479142561354364e-05,
|
| 245 |
+
"loss": 0.612,
|
| 246 |
+
"step": 16000
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 1.3564616902334758,
|
| 250 |
+
"grad_norm": 2.6160857677459717,
|
| 251 |
+
"learning_rate": 1.217465440594361e-05,
|
| 252 |
+
"loss": 0.6179,
|
| 253 |
+
"step": 16500
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 1.3975665899375205,
|
| 257 |
+
"grad_norm": 3.3576176166534424,
|
| 258 |
+
"learning_rate": 1.1870166250532855e-05,
|
| 259 |
+
"loss": 0.6203,
|
| 260 |
+
"step": 17000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 1.4386714896415653,
|
| 264 |
+
"grad_norm": 2.6305747032165527,
|
| 265 |
+
"learning_rate": 1.1565678095122101e-05,
|
| 266 |
+
"loss": 0.609,
|
| 267 |
+
"step": 17500
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 1.47977638934561,
|
| 271 |
+
"grad_norm": 3.3986129760742188,
|
| 272 |
+
"learning_rate": 1.1261189939711345e-05,
|
| 273 |
+
"loss": 0.6076,
|
| 274 |
+
"step": 18000
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 1.5208812890496546,
|
| 278 |
+
"grad_norm": 5.06044864654541,
|
| 279 |
+
"learning_rate": 1.0956701784300592e-05,
|
| 280 |
+
"loss": 0.6187,
|
| 281 |
+
"step": 18500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 1.5619861887536994,
|
| 285 |
+
"grad_norm": 4.3670549392700195,
|
| 286 |
+
"learning_rate": 1.0652213628889839e-05,
|
| 287 |
+
"loss": 0.6063,
|
| 288 |
+
"step": 19000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 1.6030910884577443,
|
| 292 |
+
"grad_norm": 8.928343772888184,
|
| 293 |
+
"learning_rate": 1.0347725473479082e-05,
|
| 294 |
+
"loss": 0.5948,
|
| 295 |
+
"step": 19500
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 1.644195988161789,
|
| 299 |
+
"grad_norm": 3.5316221714019775,
|
| 300 |
+
"learning_rate": 1.004323731806833e-05,
|
| 301 |
+
"loss": 0.6108,
|
| 302 |
+
"step": 20000
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 1.6853008878658335,
|
| 306 |
+
"grad_norm": 3.8091230392456055,
|
| 307 |
+
"learning_rate": 9.738749162657574e-06,
|
| 308 |
+
"loss": 0.6024,
|
| 309 |
+
"step": 20500
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.7264057875698784,
|
| 313 |
+
"grad_norm": 2.5065314769744873,
|
| 314 |
+
"learning_rate": 9.43426100724682e-06,
|
| 315 |
+
"loss": 0.595,
|
| 316 |
+
"step": 21000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.767510687273923,
|
| 320 |
+
"grad_norm": 4.371850490570068,
|
| 321 |
+
"learning_rate": 9.129772851836063e-06,
|
| 322 |
+
"loss": 0.6063,
|
| 323 |
+
"step": 21500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 1.8086155869779676,
|
| 327 |
+
"grad_norm": 3.6098592281341553,
|
| 328 |
+
"learning_rate": 8.82528469642531e-06,
|
| 329 |
+
"loss": 0.6158,
|
| 330 |
+
"step": 22000
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 1.8497204866820125,
|
| 334 |
+
"grad_norm": 4.037623405456543,
|
| 335 |
+
"learning_rate": 8.520796541014555e-06,
|
| 336 |
+
"loss": 0.6045,
|
| 337 |
+
"step": 22500
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 1.8908253863860573,
|
| 341 |
+
"grad_norm": 4.389125823974609,
|
| 342 |
+
"learning_rate": 8.2163083856038e-06,
|
| 343 |
+
"loss": 0.5876,
|
| 344 |
+
"step": 23000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 1.931930286090102,
|
| 348 |
+
"grad_norm": 4.564250946044922,
|
| 349 |
+
"learning_rate": 7.911820230193046e-06,
|
| 350 |
+
"loss": 0.5857,
|
| 351 |
+
"step": 23500
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 1.9730351857941466,
|
| 355 |
+
"grad_norm": 4.7007365226745605,
|
| 356 |
+
"learning_rate": 7.607332074782292e-06,
|
| 357 |
+
"loss": 0.6188,
|
| 358 |
+
"step": 24000
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 2.0,
|
| 362 |
+
"eval_gen_len": 34.1084,
|
| 363 |
+
"eval_loss": 0.5606569051742554,
|
| 364 |
+
"eval_rouge1": 32.2894,
|
| 365 |
+
"eval_rouge2": 27.8387,
|
| 366 |
+
"eval_rougeL": 31.7251,
|
| 367 |
+
"eval_rougeLsum": 31.8152,
|
| 368 |
+
"eval_runtime": 1524.3799,
|
| 369 |
+
"eval_samples_per_second": 14.185,
|
| 370 |
+
"eval_steps_per_second": 0.443,
|
| 371 |
+
"step": 24328
|
| 372 |
+
}
|
| 373 |
+
],
|
| 374 |
+
"logging_steps": 500,
|
| 375 |
+
"max_steps": 36492,
|
| 376 |
+
"num_input_tokens_seen": 0,
|
| 377 |
+
"num_train_epochs": 3,
|
| 378 |
+
"save_steps": 500,
|
| 379 |
+
"stateful_callbacks": {
|
| 380 |
+
"EarlyStoppingCallback": {
|
| 381 |
+
"args": {
|
| 382 |
+
"early_stopping_patience": 5,
|
| 383 |
+
"early_stopping_threshold": 0.01
|
| 384 |
+
},
|
| 385 |
+
"attributes": {
|
| 386 |
+
"early_stopping_patience_counter": 0
|
| 387 |
+
}
|
| 388 |
+
},
|
| 389 |
+
"TrainerControl": {
|
| 390 |
+
"args": {
|
| 391 |
+
"should_epoch_stop": false,
|
| 392 |
+
"should_evaluate": false,
|
| 393 |
+
"should_log": false,
|
| 394 |
+
"should_save": true,
|
| 395 |
+
"should_training_stop": false
|
| 396 |
+
},
|
| 397 |
+
"attributes": {}
|
| 398 |
+
}
|
| 399 |
+
},
|
| 400 |
+
"total_flos": 8379853808074752.0,
|
| 401 |
+
"train_batch_size": 8,
|
| 402 |
+
"trial_name": null,
|
| 403 |
+
"trial_params": null
|
| 404 |
+
}
|
checkpoint-24328/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7b6cdaabc5186df4c830688bac90d2f77a2412e2b54715163a34f977335aa8ac
|
| 3 |
+
size 5496
|
checkpoint-24328/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-36492/config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Helsinki-NLP/opus-mt-en-ru",
|
| 3 |
+
"_num_labels": 3,
|
| 4 |
+
"activation_dropout": 0.0,
|
| 5 |
+
"activation_function": "swish",
|
| 6 |
+
"add_bias_logits": false,
|
| 7 |
+
"add_final_layer_norm": false,
|
| 8 |
+
"architectures": [
|
| 9 |
+
"MarianMTModel"
|
| 10 |
+
],
|
| 11 |
+
"attention_dropout": 0.0,
|
| 12 |
+
"bos_token_id": 0,
|
| 13 |
+
"classif_dropout": 0.0,
|
| 14 |
+
"classifier_dropout": 0.0,
|
| 15 |
+
"d_model": 512,
|
| 16 |
+
"decoder_attention_heads": 8,
|
| 17 |
+
"decoder_ffn_dim": 2048,
|
| 18 |
+
"decoder_layerdrop": 0.0,
|
| 19 |
+
"decoder_layers": 6,
|
| 20 |
+
"decoder_start_token_id": 62517,
|
| 21 |
+
"decoder_vocab_size": 62518,
|
| 22 |
+
"dropout": 0.1,
|
| 23 |
+
"encoder_attention_heads": 8,
|
| 24 |
+
"encoder_ffn_dim": 2048,
|
| 25 |
+
"encoder_layerdrop": 0.0,
|
| 26 |
+
"encoder_layers": 6,
|
| 27 |
+
"eos_token_id": 0,
|
| 28 |
+
"forced_eos_token_id": 0,
|
| 29 |
+
"id2label": {
|
| 30 |
+
"0": "LABEL_0",
|
| 31 |
+
"1": "LABEL_1",
|
| 32 |
+
"2": "LABEL_2"
|
| 33 |
+
},
|
| 34 |
+
"init_std": 0.02,
|
| 35 |
+
"is_encoder_decoder": true,
|
| 36 |
+
"label2id": {
|
| 37 |
+
"LABEL_0": 0,
|
| 38 |
+
"LABEL_1": 1,
|
| 39 |
+
"LABEL_2": 2
|
| 40 |
+
},
|
| 41 |
+
"max_length": null,
|
| 42 |
+
"max_position_embeddings": 512,
|
| 43 |
+
"model_type": "marian",
|
| 44 |
+
"normalize_before": false,
|
| 45 |
+
"normalize_embedding": false,
|
| 46 |
+
"num_beams": null,
|
| 47 |
+
"num_hidden_layers": 6,
|
| 48 |
+
"pad_token_id": 62517,
|
| 49 |
+
"scale_embedding": true,
|
| 50 |
+
"share_encoder_decoder_embeddings": true,
|
| 51 |
+
"static_position_embeddings": true,
|
| 52 |
+
"torch_dtype": "float32",
|
| 53 |
+
"transformers_version": "4.48.0",
|
| 54 |
+
"use_cache": false,
|
| 55 |
+
"vocab_size": 62518
|
| 56 |
+
}
|
checkpoint-36492/generation_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bad_words_ids": [
|
| 3 |
+
[
|
| 4 |
+
62517
|
| 5 |
+
]
|
| 6 |
+
],
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"decoder_start_token_id": 62517,
|
| 9 |
+
"eos_token_id": 0,
|
| 10 |
+
"forced_eos_token_id": 0,
|
| 11 |
+
"max_length": 512,
|
| 12 |
+
"num_beams": 4,
|
| 13 |
+
"pad_token_id": 62517,
|
| 14 |
+
"renormalize_logits": true,
|
| 15 |
+
"transformers_version": "4.48.0"
|
| 16 |
+
}
|
checkpoint-36492/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:299782ae957802fbb6326a979be87baa6e1af0752fc628ad1b8603565ba98c82
|
| 3 |
+
size 304869976
|
checkpoint-36492/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:43073178bdc78381cd9e75014954229a2d8ce3961fa0d6f5530c2d3e02123e67
|
| 3 |
+
size 609393274
|
checkpoint-36492/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:feb2244b9e3c379b21e267dc0e4b5b119f0584d6561dba538bd6d64835907f19
|
| 3 |
+
size 14512
|
checkpoint-36492/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:28f36686ed5b4af5f2fc7e7972a2d7bd91bbf27fdaf653759297e9ef53d9e57d
|
| 3 |
+
size 14512
|
checkpoint-36492/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1f6550b50dab793d8df12535663f382e792fe37fddac1a48240ad18ddaa075d6
|
| 3 |
+
size 1064
|
checkpoint-36492/source.spm
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:16bebef1389a0b8ab452772c4e35b9e605e5713f8ac7baa71ca701394eaa086d
|
| 3 |
+
size 802781
|
checkpoint-36492/special_tokens_map.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eos_token": "</s>",
|
| 3 |
+
"pad_token": "<pad>",
|
| 4 |
+
"unk_token": "<unk>"
|
| 5 |
+
}
|
checkpoint-36492/target.spm
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:745998e51ba5b058e38b7ac7765c25c43ed5c1c39cc92b27163b9b2e323c9d7c
|
| 3 |
+
size 1080169
|
checkpoint-36492/tokenizer_config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "<unk>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"62517": {
|
| 20 |
+
"content": "<pad>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
}
|
| 27 |
+
},
|
| 28 |
+
"clean_up_tokenization_spaces": false,
|
| 29 |
+
"eos_token": "</s>",
|
| 30 |
+
"extra_special_tokens": {},
|
| 31 |
+
"model_max_length": 512,
|
| 32 |
+
"pad_token": "<pad>",
|
| 33 |
+
"separate_vocabs": false,
|
| 34 |
+
"source_lang": "en",
|
| 35 |
+
"sp_model_kwargs": {},
|
| 36 |
+
"target_lang": "ru",
|
| 37 |
+
"tokenizer_class": "MarianTokenizer",
|
| 38 |
+
"unk_token": "<unk>"
|
| 39 |
+
}
|
checkpoint-36492/trainer_state.json
ADDED
|
@@ -0,0 +1,585 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.5498164296150208,
|
| 3 |
+
"best_model_checkpoint": "ck3-localization/checkpoint-36492",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 36492,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.041104899704044726,
|
| 13 |
+
"grad_norm": 6.12867546081543,
|
| 14 |
+
"learning_rate": 2.7397260273972604e-06,
|
| 15 |
+
"loss": 2.3513,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.08220979940808945,
|
| 20 |
+
"grad_norm": 6.480109691619873,
|
| 21 |
+
"learning_rate": 5.479452054794521e-06,
|
| 22 |
+
"loss": 1.4689,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.12331469911213416,
|
| 27 |
+
"grad_norm": 4.740694046020508,
|
| 28 |
+
"learning_rate": 8.219178082191782e-06,
|
| 29 |
+
"loss": 1.19,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.1644195988161789,
|
| 34 |
+
"grad_norm": 5.501825332641602,
|
| 35 |
+
"learning_rate": 1.0958904109589042e-05,
|
| 36 |
+
"loss": 1.0318,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.20552449852022361,
|
| 41 |
+
"grad_norm": 5.322640419006348,
|
| 42 |
+
"learning_rate": 1.3698630136986302e-05,
|
| 43 |
+
"loss": 0.9556,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.24662939822426833,
|
| 48 |
+
"grad_norm": 5.773024559020996,
|
| 49 |
+
"learning_rate": 1.6438356164383563e-05,
|
| 50 |
+
"loss": 0.9137,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.28773429792831307,
|
| 55 |
+
"grad_norm": 4.7130656242370605,
|
| 56 |
+
"learning_rate": 1.9178082191780822e-05,
|
| 57 |
+
"loss": 0.8705,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.3288391976323578,
|
| 62 |
+
"grad_norm": 3.756901741027832,
|
| 63 |
+
"learning_rate": 1.9786858291212475e-05,
|
| 64 |
+
"loss": 0.8531,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.3699440973364025,
|
| 69 |
+
"grad_norm": 4.033214092254639,
|
| 70 |
+
"learning_rate": 1.948237013580172e-05,
|
| 71 |
+
"loss": 0.8138,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.41104899704044723,
|
| 76 |
+
"grad_norm": 4.564835071563721,
|
| 77 |
+
"learning_rate": 1.9177881980390966e-05,
|
| 78 |
+
"loss": 0.7946,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.45215389674449197,
|
| 83 |
+
"grad_norm": 4.850255012512207,
|
| 84 |
+
"learning_rate": 1.887339382498021e-05,
|
| 85 |
+
"loss": 0.7833,
|
| 86 |
+
"step": 5500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.49325879644853665,
|
| 90 |
+
"grad_norm": 4.716800689697266,
|
| 91 |
+
"learning_rate": 1.8568905669569456e-05,
|
| 92 |
+
"loss": 0.7662,
|
| 93 |
+
"step": 6000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.5343636961525814,
|
| 97 |
+
"grad_norm": 4.0613274574279785,
|
| 98 |
+
"learning_rate": 1.82644175141587e-05,
|
| 99 |
+
"loss": 0.7668,
|
| 100 |
+
"step": 6500
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.5754685958566261,
|
| 104 |
+
"grad_norm": 4.6606831550598145,
|
| 105 |
+
"learning_rate": 1.7959929358747943e-05,
|
| 106 |
+
"loss": 0.7518,
|
| 107 |
+
"step": 7000
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.6165734955606709,
|
| 111 |
+
"grad_norm": 4.745144844055176,
|
| 112 |
+
"learning_rate": 1.765544120333719e-05,
|
| 113 |
+
"loss": 0.7477,
|
| 114 |
+
"step": 7500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.6576783952647156,
|
| 118 |
+
"grad_norm": 4.414773464202881,
|
| 119 |
+
"learning_rate": 1.7350953047926437e-05,
|
| 120 |
+
"loss": 0.751,
|
| 121 |
+
"step": 8000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.6987832949687602,
|
| 125 |
+
"grad_norm": 4.531556606292725,
|
| 126 |
+
"learning_rate": 1.704646489251568e-05,
|
| 127 |
+
"loss": 0.7424,
|
| 128 |
+
"step": 8500
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.739888194672805,
|
| 132 |
+
"grad_norm": 4.841054439544678,
|
| 133 |
+
"learning_rate": 1.6741976737104928e-05,
|
| 134 |
+
"loss": 0.7328,
|
| 135 |
+
"step": 9000
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.7809930943768497,
|
| 139 |
+
"grad_norm": 4.302633762359619,
|
| 140 |
+
"learning_rate": 1.6437488581694175e-05,
|
| 141 |
+
"loss": 0.725,
|
| 142 |
+
"step": 9500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.8220979940808945,
|
| 146 |
+
"grad_norm": 3.530836343765259,
|
| 147 |
+
"learning_rate": 1.6133000426283418e-05,
|
| 148 |
+
"loss": 0.7072,
|
| 149 |
+
"step": 10000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.8632028937849392,
|
| 153 |
+
"grad_norm": 4.2016191482543945,
|
| 154 |
+
"learning_rate": 1.5828512270872665e-05,
|
| 155 |
+
"loss": 0.6895,
|
| 156 |
+
"step": 10500
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.9043077934889839,
|
| 160 |
+
"grad_norm": 3.463226079940796,
|
| 161 |
+
"learning_rate": 1.5524024115461912e-05,
|
| 162 |
+
"loss": 0.7093,
|
| 163 |
+
"step": 11000
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.9454126931930286,
|
| 167 |
+
"grad_norm": 3.3261220455169678,
|
| 168 |
+
"learning_rate": 1.5219535960051155e-05,
|
| 169 |
+
"loss": 0.6941,
|
| 170 |
+
"step": 11500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.9865175928970733,
|
| 174 |
+
"grad_norm": 3.778331756591797,
|
| 175 |
+
"learning_rate": 1.49150478046404e-05,
|
| 176 |
+
"loss": 0.7008,
|
| 177 |
+
"step": 12000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.0,
|
| 181 |
+
"eval_gen_len": 33.8514,
|
| 182 |
+
"eval_loss": 0.6079365611076355,
|
| 183 |
+
"eval_rouge1": 31.9733,
|
| 184 |
+
"eval_rouge2": 27.2686,
|
| 185 |
+
"eval_rougeL": 31.3526,
|
| 186 |
+
"eval_rougeLsum": 31.4543,
|
| 187 |
+
"eval_runtime": 1591.2259,
|
| 188 |
+
"eval_samples_per_second": 13.59,
|
| 189 |
+
"eval_steps_per_second": 0.425,
|
| 190 |
+
"step": 12164
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 1.0276224926011182,
|
| 194 |
+
"grad_norm": 3.7499120235443115,
|
| 195 |
+
"learning_rate": 1.4610559649229646e-05,
|
| 196 |
+
"loss": 0.6284,
|
| 197 |
+
"step": 12500
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 1.0687273923051628,
|
| 201 |
+
"grad_norm": 3.52150821685791,
|
| 202 |
+
"learning_rate": 1.4306071493818891e-05,
|
| 203 |
+
"loss": 0.6337,
|
| 204 |
+
"step": 13000
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 1.1098322920092074,
|
| 208 |
+
"grad_norm": 4.520621299743652,
|
| 209 |
+
"learning_rate": 1.4001583338408138e-05,
|
| 210 |
+
"loss": 0.6292,
|
| 211 |
+
"step": 13500
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 1.1509371917132523,
|
| 215 |
+
"grad_norm": 4.275709629058838,
|
| 216 |
+
"learning_rate": 1.3697095182997382e-05,
|
| 217 |
+
"loss": 0.6195,
|
| 218 |
+
"step": 14000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 1.192042091417297,
|
| 222 |
+
"grad_norm": 3.500743865966797,
|
| 223 |
+
"learning_rate": 1.3392607027586628e-05,
|
| 224 |
+
"loss": 0.6168,
|
| 225 |
+
"step": 14500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 1.2331469911213417,
|
| 229 |
+
"grad_norm": 2.587315320968628,
|
| 230 |
+
"learning_rate": 1.3088118872175875e-05,
|
| 231 |
+
"loss": 0.6169,
|
| 232 |
+
"step": 15000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 1.2742518908253864,
|
| 236 |
+
"grad_norm": 3.4539663791656494,
|
| 237 |
+
"learning_rate": 1.2783630716765119e-05,
|
| 238 |
+
"loss": 0.619,
|
| 239 |
+
"step": 15500
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 1.3153567905294312,
|
| 243 |
+
"grad_norm": 3.759758710861206,
|
| 244 |
+
"learning_rate": 1.2479142561354364e-05,
|
| 245 |
+
"loss": 0.612,
|
| 246 |
+
"step": 16000
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 1.3564616902334758,
|
| 250 |
+
"grad_norm": 2.6160857677459717,
|
| 251 |
+
"learning_rate": 1.217465440594361e-05,
|
| 252 |
+
"loss": 0.6179,
|
| 253 |
+
"step": 16500
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 1.3975665899375205,
|
| 257 |
+
"grad_norm": 3.3576176166534424,
|
| 258 |
+
"learning_rate": 1.1870166250532855e-05,
|
| 259 |
+
"loss": 0.6203,
|
| 260 |
+
"step": 17000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 1.4386714896415653,
|
| 264 |
+
"grad_norm": 2.6305747032165527,
|
| 265 |
+
"learning_rate": 1.1565678095122101e-05,
|
| 266 |
+
"loss": 0.609,
|
| 267 |
+
"step": 17500
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 1.47977638934561,
|
| 271 |
+
"grad_norm": 3.3986129760742188,
|
| 272 |
+
"learning_rate": 1.1261189939711345e-05,
|
| 273 |
+
"loss": 0.6076,
|
| 274 |
+
"step": 18000
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 1.5208812890496546,
|
| 278 |
+
"grad_norm": 5.06044864654541,
|
| 279 |
+
"learning_rate": 1.0956701784300592e-05,
|
| 280 |
+
"loss": 0.6187,
|
| 281 |
+
"step": 18500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 1.5619861887536994,
|
| 285 |
+
"grad_norm": 4.3670549392700195,
|
| 286 |
+
"learning_rate": 1.0652213628889839e-05,
|
| 287 |
+
"loss": 0.6063,
|
| 288 |
+
"step": 19000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 1.6030910884577443,
|
| 292 |
+
"grad_norm": 8.928343772888184,
|
| 293 |
+
"learning_rate": 1.0347725473479082e-05,
|
| 294 |
+
"loss": 0.5948,
|
| 295 |
+
"step": 19500
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 1.644195988161789,
|
| 299 |
+
"grad_norm": 3.5316221714019775,
|
| 300 |
+
"learning_rate": 1.004323731806833e-05,
|
| 301 |
+
"loss": 0.6108,
|
| 302 |
+
"step": 20000
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 1.6853008878658335,
|
| 306 |
+
"grad_norm": 3.8091230392456055,
|
| 307 |
+
"learning_rate": 9.738749162657574e-06,
|
| 308 |
+
"loss": 0.6024,
|
| 309 |
+
"step": 20500
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.7264057875698784,
|
| 313 |
+
"grad_norm": 2.5065314769744873,
|
| 314 |
+
"learning_rate": 9.43426100724682e-06,
|
| 315 |
+
"loss": 0.595,
|
| 316 |
+
"step": 21000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.767510687273923,
|
| 320 |
+
"grad_norm": 4.371850490570068,
|
| 321 |
+
"learning_rate": 9.129772851836063e-06,
|
| 322 |
+
"loss": 0.6063,
|
| 323 |
+
"step": 21500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 1.8086155869779676,
|
| 327 |
+
"grad_norm": 3.6098592281341553,
|
| 328 |
+
"learning_rate": 8.82528469642531e-06,
|
| 329 |
+
"loss": 0.6158,
|
| 330 |
+
"step": 22000
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 1.8497204866820125,
|
| 334 |
+
"grad_norm": 4.037623405456543,
|
| 335 |
+
"learning_rate": 8.520796541014555e-06,
|
| 336 |
+
"loss": 0.6045,
|
| 337 |
+
"step": 22500
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 1.8908253863860573,
|
| 341 |
+
"grad_norm": 4.389125823974609,
|
| 342 |
+
"learning_rate": 8.2163083856038e-06,
|
| 343 |
+
"loss": 0.5876,
|
| 344 |
+
"step": 23000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 1.931930286090102,
|
| 348 |
+
"grad_norm": 4.564250946044922,
|
| 349 |
+
"learning_rate": 7.911820230193046e-06,
|
| 350 |
+
"loss": 0.5857,
|
| 351 |
+
"step": 23500
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 1.9730351857941466,
|
| 355 |
+
"grad_norm": 4.7007365226745605,
|
| 356 |
+
"learning_rate": 7.607332074782292e-06,
|
| 357 |
+
"loss": 0.6188,
|
| 358 |
+
"step": 24000
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 2.0,
|
| 362 |
+
"eval_gen_len": 34.1084,
|
| 363 |
+
"eval_loss": 0.5606569051742554,
|
| 364 |
+
"eval_rouge1": 32.2894,
|
| 365 |
+
"eval_rouge2": 27.8387,
|
| 366 |
+
"eval_rougeL": 31.7251,
|
| 367 |
+
"eval_rougeLsum": 31.8152,
|
| 368 |
+
"eval_runtime": 1524.3799,
|
| 369 |
+
"eval_samples_per_second": 14.185,
|
| 370 |
+
"eval_steps_per_second": 0.443,
|
| 371 |
+
"step": 24328
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 2.0141400854981915,
|
| 375 |
+
"grad_norm": 1.9762619733810425,
|
| 376 |
+
"learning_rate": 7.302843919371537e-06,
|
| 377 |
+
"loss": 0.5736,
|
| 378 |
+
"step": 24500
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 2.0552449852022363,
|
| 382 |
+
"grad_norm": 4.161807060241699,
|
| 383 |
+
"learning_rate": 6.998355763960782e-06,
|
| 384 |
+
"loss": 0.5458,
|
| 385 |
+
"step": 25000
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 2.0963498849062807,
|
| 389 |
+
"grad_norm": 2.221466541290283,
|
| 390 |
+
"learning_rate": 6.693867608550027e-06,
|
| 391 |
+
"loss": 0.5468,
|
| 392 |
+
"step": 25500
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 2.1374547846103256,
|
| 396 |
+
"grad_norm": 3.642603874206543,
|
| 397 |
+
"learning_rate": 6.3893794531392735e-06,
|
| 398 |
+
"loss": 0.5556,
|
| 399 |
+
"step": 26000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 2.1785596843143704,
|
| 403 |
+
"grad_norm": 3.3420753479003906,
|
| 404 |
+
"learning_rate": 6.084891297728519e-06,
|
| 405 |
+
"loss": 0.5451,
|
| 406 |
+
"step": 26500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 2.219664584018415,
|
| 410 |
+
"grad_norm": 3.2656307220458984,
|
| 411 |
+
"learning_rate": 5.780403142317764e-06,
|
| 412 |
+
"loss": 0.5303,
|
| 413 |
+
"step": 27000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 2.2607694837224597,
|
| 417 |
+
"grad_norm": 3.4211599826812744,
|
| 418 |
+
"learning_rate": 5.475914986907009e-06,
|
| 419 |
+
"loss": 0.5496,
|
| 420 |
+
"step": 27500
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 2.3018743834265045,
|
| 424 |
+
"grad_norm": 4.965065956115723,
|
| 425 |
+
"learning_rate": 5.171426831496256e-06,
|
| 426 |
+
"loss": 0.5392,
|
| 427 |
+
"step": 28000
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 2.342979283130549,
|
| 431 |
+
"grad_norm": 2.4368653297424316,
|
| 432 |
+
"learning_rate": 4.866938676085501e-06,
|
| 433 |
+
"loss": 0.5443,
|
| 434 |
+
"step": 28500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 2.384084182834594,
|
| 438 |
+
"grad_norm": 2.6363422870635986,
|
| 439 |
+
"learning_rate": 4.562450520674746e-06,
|
| 440 |
+
"loss": 0.5418,
|
| 441 |
+
"step": 29000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 2.4251890825386386,
|
| 445 |
+
"grad_norm": 4.417261123657227,
|
| 446 |
+
"learning_rate": 4.257962365263992e-06,
|
| 447 |
+
"loss": 0.5377,
|
| 448 |
+
"step": 29500
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 2.4662939822426835,
|
| 452 |
+
"grad_norm": 2.850874900817871,
|
| 453 |
+
"learning_rate": 3.953474209853237e-06,
|
| 454 |
+
"loss": 0.5406,
|
| 455 |
+
"step": 30000
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 2.507398881946728,
|
| 459 |
+
"grad_norm": 3.0470714569091797,
|
| 460 |
+
"learning_rate": 3.6489860544424825e-06,
|
| 461 |
+
"loss": 0.5277,
|
| 462 |
+
"step": 30500
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 2.5485037816507727,
|
| 466 |
+
"grad_norm": 3.77066707611084,
|
| 467 |
+
"learning_rate": 3.3444978990317278e-06,
|
| 468 |
+
"loss": 0.5404,
|
| 469 |
+
"step": 31000
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 2.5896086813548176,
|
| 473 |
+
"grad_norm": 2.103886127471924,
|
| 474 |
+
"learning_rate": 3.0400097436209734e-06,
|
| 475 |
+
"loss": 0.5511,
|
| 476 |
+
"step": 31500
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 2.6307135810588624,
|
| 480 |
+
"grad_norm": 2.784128427505493,
|
| 481 |
+
"learning_rate": 2.7355215882102186e-06,
|
| 482 |
+
"loss": 0.5466,
|
| 483 |
+
"step": 32000
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 2.671818480762907,
|
| 487 |
+
"grad_norm": 2.9157445430755615,
|
| 488 |
+
"learning_rate": 2.4310334327994643e-06,
|
| 489 |
+
"loss": 0.5502,
|
| 490 |
+
"step": 32500
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 2.7129233804669517,
|
| 494 |
+
"grad_norm": 3.7201719284057617,
|
| 495 |
+
"learning_rate": 2.1265452773887095e-06,
|
| 496 |
+
"loss": 0.5435,
|
| 497 |
+
"step": 33000
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 2.7540282801709965,
|
| 501 |
+
"grad_norm": 3.0712995529174805,
|
| 502 |
+
"learning_rate": 1.8220571219779551e-06,
|
| 503 |
+
"loss": 0.5442,
|
| 504 |
+
"step": 33500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 2.795133179875041,
|
| 508 |
+
"grad_norm": 3.5120229721069336,
|
| 509 |
+
"learning_rate": 1.5175689665672005e-06,
|
| 510 |
+
"loss": 0.5392,
|
| 511 |
+
"step": 34000
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 2.836238079579086,
|
| 515 |
+
"grad_norm": 3.09930682182312,
|
| 516 |
+
"learning_rate": 1.2130808111564462e-06,
|
| 517 |
+
"loss": 0.5355,
|
| 518 |
+
"step": 34500
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 2.8773429792831307,
|
| 522 |
+
"grad_norm": 2.3439698219299316,
|
| 523 |
+
"learning_rate": 9.085926557456916e-07,
|
| 524 |
+
"loss": 0.5531,
|
| 525 |
+
"step": 35000
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 2.918447878987175,
|
| 529 |
+
"grad_norm": 2.935701608657837,
|
| 530 |
+
"learning_rate": 6.04104500334937e-07,
|
| 531 |
+
"loss": 0.5355,
|
| 532 |
+
"step": 35500
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 2.95955277869122,
|
| 536 |
+
"grad_norm": 3.3096513748168945,
|
| 537 |
+
"learning_rate": 2.9961634492418246e-07,
|
| 538 |
+
"loss": 0.5528,
|
| 539 |
+
"step": 36000
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 3.0,
|
| 543 |
+
"eval_gen_len": 33.8306,
|
| 544 |
+
"eval_loss": 0.5498164296150208,
|
| 545 |
+
"eval_rouge1": 32.6478,
|
| 546 |
+
"eval_rouge2": 28.3022,
|
| 547 |
+
"eval_rougeL": 32.1001,
|
| 548 |
+
"eval_rougeLsum": 32.1838,
|
| 549 |
+
"eval_runtime": 1535.5456,
|
| 550 |
+
"eval_samples_per_second": 14.082,
|
| 551 |
+
"eval_steps_per_second": 0.44,
|
| 552 |
+
"step": 36492
|
| 553 |
+
}
|
| 554 |
+
],
|
| 555 |
+
"logging_steps": 500,
|
| 556 |
+
"max_steps": 36492,
|
| 557 |
+
"num_input_tokens_seen": 0,
|
| 558 |
+
"num_train_epochs": 3,
|
| 559 |
+
"save_steps": 500,
|
| 560 |
+
"stateful_callbacks": {
|
| 561 |
+
"EarlyStoppingCallback": {
|
| 562 |
+
"args": {
|
| 563 |
+
"early_stopping_patience": 5,
|
| 564 |
+
"early_stopping_threshold": 0.01
|
| 565 |
+
},
|
| 566 |
+
"attributes": {
|
| 567 |
+
"early_stopping_patience_counter": 0
|
| 568 |
+
}
|
| 569 |
+
},
|
| 570 |
+
"TrainerControl": {
|
| 571 |
+
"args": {
|
| 572 |
+
"should_epoch_stop": false,
|
| 573 |
+
"should_evaluate": false,
|
| 574 |
+
"should_log": false,
|
| 575 |
+
"should_save": true,
|
| 576 |
+
"should_training_stop": true
|
| 577 |
+
},
|
| 578 |
+
"attributes": {}
|
| 579 |
+
}
|
| 580 |
+
},
|
| 581 |
+
"total_flos": 1.2577075189776384e+16,
|
| 582 |
+
"train_batch_size": 8,
|
| 583 |
+
"trial_name": null,
|
| 584 |
+
"trial_params": null
|
| 585 |
+
}
|
checkpoint-36492/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7b6cdaabc5186df4c830688bac90d2f77a2412e2b54715163a34f977335aa8ac
|
| 3 |
+
size 5496
|
checkpoint-36492/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Helsinki-NLP/opus-mt-en-ru",
|
| 3 |
+
"_num_labels": 3,
|
| 4 |
+
"activation_dropout": 0.0,
|
| 5 |
+
"activation_function": "swish",
|
| 6 |
+
"add_bias_logits": false,
|
| 7 |
+
"add_final_layer_norm": false,
|
| 8 |
+
"architectures": [
|
| 9 |
+
"MarianMTModel"
|
| 10 |
+
],
|
| 11 |
+
"attention_dropout": 0.0,
|
| 12 |
+
"bos_token_id": 0,
|
| 13 |
+
"classif_dropout": 0.0,
|
| 14 |
+
"classifier_dropout": 0.0,
|
| 15 |
+
"d_model": 512,
|
| 16 |
+
"decoder_attention_heads": 8,
|
| 17 |
+
"decoder_ffn_dim": 2048,
|
| 18 |
+
"decoder_layerdrop": 0.0,
|
| 19 |
+
"decoder_layers": 6,
|
| 20 |
+
"decoder_start_token_id": 62517,
|
| 21 |
+
"decoder_vocab_size": 62518,
|
| 22 |
+
"dropout": 0.1,
|
| 23 |
+
"encoder_attention_heads": 8,
|
| 24 |
+
"encoder_ffn_dim": 2048,
|
| 25 |
+
"encoder_layerdrop": 0.0,
|
| 26 |
+
"encoder_layers": 6,
|
| 27 |
+
"eos_token_id": 0,
|
| 28 |
+
"forced_eos_token_id": 0,
|
| 29 |
+
"id2label": {
|
| 30 |
+
"0": "LABEL_0",
|
| 31 |
+
"1": "LABEL_1",
|
| 32 |
+
"2": "LABEL_2"
|
| 33 |
+
},
|
| 34 |
+
"init_std": 0.02,
|
| 35 |
+
"is_encoder_decoder": true,
|
| 36 |
+
"label2id": {
|
| 37 |
+
"LABEL_0": 0,
|
| 38 |
+
"LABEL_1": 1,
|
| 39 |
+
"LABEL_2": 2
|
| 40 |
+
},
|
| 41 |
+
"max_length": null,
|
| 42 |
+
"max_position_embeddings": 512,
|
| 43 |
+
"model_type": "marian",
|
| 44 |
+
"normalize_before": false,
|
| 45 |
+
"normalize_embedding": false,
|
| 46 |
+
"num_beams": null,
|
| 47 |
+
"num_hidden_layers": 6,
|
| 48 |
+
"pad_token_id": 62517,
|
| 49 |
+
"scale_embedding": true,
|
| 50 |
+
"share_encoder_decoder_embeddings": true,
|
| 51 |
+
"static_position_embeddings": true,
|
| 52 |
+
"torch_dtype": "float32",
|
| 53 |
+
"transformers_version": "4.48.0",
|
| 54 |
+
"use_cache": true,
|
| 55 |
+
"vocab_size": 62518
|
| 56 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bad_words_ids": [
|
| 3 |
+
[
|
| 4 |
+
62517
|
| 5 |
+
]
|
| 6 |
+
],
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"decoder_start_token_id": 62517,
|
| 9 |
+
"eos_token_id": 0,
|
| 10 |
+
"forced_eos_token_id": 0,
|
| 11 |
+
"max_length": 512,
|
| 12 |
+
"num_beams": 4,
|
| 13 |
+
"pad_token_id": 62517,
|
| 14 |
+
"renormalize_logits": true,
|
| 15 |
+
"transformers_version": "4.48.0"
|
| 16 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:299782ae957802fbb6326a979be87baa6e1af0752fc628ad1b8603565ba98c82
|
| 3 |
+
size 304869976
|
source.spm
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:16bebef1389a0b8ab452772c4e35b9e605e5713f8ac7baa71ca701394eaa086d
|
| 3 |
+
size 802781
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eos_token": "</s>",
|
| 3 |
+
"pad_token": "<pad>",
|
| 4 |
+
"unk_token": "<unk>"
|
| 5 |
+
}
|
target.spm
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:745998e51ba5b058e38b7ac7765c25c43ed5c1c39cc92b27163b9b2e323c9d7c
|
| 3 |
+
size 1080169
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "<unk>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"62517": {
|
| 20 |
+
"content": "<pad>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
}
|
| 27 |
+
},
|
| 28 |
+
"clean_up_tokenization_spaces": false,
|
| 29 |
+
"eos_token": "</s>",
|
| 30 |
+
"extra_special_tokens": {},
|
| 31 |
+
"model_max_length": 512,
|
| 32 |
+
"pad_token": "<pad>",
|
| 33 |
+
"separate_vocabs": false,
|
| 34 |
+
"source_lang": "en",
|
| 35 |
+
"sp_model_kwargs": {},
|
| 36 |
+
"target_lang": "ru",
|
| 37 |
+
"tokenizer_class": "MarianTokenizer",
|
| 38 |
+
"unk_token": "<unk>"
|
| 39 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7b6cdaabc5186df4c830688bac90d2f77a2412e2b54715163a34f977335aa8ac
|
| 3 |
+
size 5496
|
training_params.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"data_path": "ck3-localization/autotrain-data",
|
| 3 |
+
"model": "Helsinki-NLP/opus-mt-en-ru",
|
| 4 |
+
"username": "Aze4ka",
|
| 5 |
+
"seed": 42,
|
| 6 |
+
"train_split": "train",
|
| 7 |
+
"valid_split": "validation",
|
| 8 |
+
"project_name": "ck3-localization",
|
| 9 |
+
"push_to_hub": true,
|
| 10 |
+
"text_column": "autotrain_text",
|
| 11 |
+
"target_column": "autotrain_label",
|
| 12 |
+
"lr": 2e-05,
|
| 13 |
+
"epochs": 3,
|
| 14 |
+
"max_seq_length": 512,
|
| 15 |
+
"max_target_length": 128,
|
| 16 |
+
"batch_size": 8,
|
| 17 |
+
"warmup_ratio": 0.1,
|
| 18 |
+
"gradient_accumulation": 1,
|
| 19 |
+
"optimizer": "adamw_torch",
|
| 20 |
+
"scheduler": "linear",
|
| 21 |
+
"weight_decay": 0.0,
|
| 22 |
+
"max_grad_norm": 1.0,
|
| 23 |
+
"logging_steps": 500,
|
| 24 |
+
"eval_strategy": "epoch",
|
| 25 |
+
"auto_find_batch_size": false,
|
| 26 |
+
"mixed_precision": null,
|
| 27 |
+
"save_total_limit": 2,
|
| 28 |
+
"peft": false,
|
| 29 |
+
"quantization": "int8",
|
| 30 |
+
"lora_r": 16,
|
| 31 |
+
"lora_alpha": 32,
|
| 32 |
+
"lora_dropout": 0.05,
|
| 33 |
+
"target_modules": "all-linear",
|
| 34 |
+
"log": "wandb",
|
| 35 |
+
"early_stopping_patience": 5,
|
| 36 |
+
"early_stopping_threshold": 0.01
|
| 37 |
+
}
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|