Upload 13 files
Browse files- config.json +61 -0
- generation_config.json +16 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- source.spm +0 -0
- special_tokens_map.json +23 -0
- target.spm +0 -0
- tokenizer_config.json +39 -0
- trainer_state.json +630 -0
- training_args.bin +3 -0
- vocab.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/home/remote/dev/llm/Helsinki-NLP/opus-mt-mul-en-word-fine-tune-en-to-pt-1k/checkpoint-3000-words-to-sentences-1k-epochs-5k-train/checkpoint-16485",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "swish",
|
| 5 |
+
"add_bias_logits": false,
|
| 6 |
+
"add_final_layer_norm": false,
|
| 7 |
+
"architectures": [
|
| 8 |
+
"MarianMTModel"
|
| 9 |
+
],
|
| 10 |
+
"attention_dropout": 0.0,
|
| 11 |
+
"bad_words_ids": [
|
| 12 |
+
[
|
| 13 |
+
64171
|
| 14 |
+
]
|
| 15 |
+
],
|
| 16 |
+
"bos_token_id": 0,
|
| 17 |
+
"classif_dropout": 0.0,
|
| 18 |
+
"classifier_dropout": 0.0,
|
| 19 |
+
"d_model": 512,
|
| 20 |
+
"decoder_attention_heads": 8,
|
| 21 |
+
"decoder_ffn_dim": 2048,
|
| 22 |
+
"decoder_layerdrop": 0.0,
|
| 23 |
+
"decoder_layers": 6,
|
| 24 |
+
"decoder_start_token_id": 64171,
|
| 25 |
+
"decoder_vocab_size": 64172,
|
| 26 |
+
"dropout": 0.1,
|
| 27 |
+
"encoder_attention_heads": 8,
|
| 28 |
+
"encoder_ffn_dim": 2048,
|
| 29 |
+
"encoder_layerdrop": 0.0,
|
| 30 |
+
"encoder_layers": 6,
|
| 31 |
+
"eos_token_id": 0,
|
| 32 |
+
"extra_pos_embeddings": 64172,
|
| 33 |
+
"forced_eos_token_id": 0,
|
| 34 |
+
"id2label": {
|
| 35 |
+
"0": "LABEL_0",
|
| 36 |
+
"1": "LABEL_1",
|
| 37 |
+
"2": "LABEL_2"
|
| 38 |
+
},
|
| 39 |
+
"init_std": 0.02,
|
| 40 |
+
"is_encoder_decoder": true,
|
| 41 |
+
"label2id": {
|
| 42 |
+
"LABEL_0": 0,
|
| 43 |
+
"LABEL_1": 1,
|
| 44 |
+
"LABEL_2": 2
|
| 45 |
+
},
|
| 46 |
+
"max_length": 512,
|
| 47 |
+
"max_position_embeddings": 512,
|
| 48 |
+
"model_type": "marian",
|
| 49 |
+
"normalize_before": false,
|
| 50 |
+
"normalize_embedding": false,
|
| 51 |
+
"num_beams": 6,
|
| 52 |
+
"num_hidden_layers": 6,
|
| 53 |
+
"pad_token_id": 64171,
|
| 54 |
+
"scale_embedding": true,
|
| 55 |
+
"share_encoder_decoder_embeddings": true,
|
| 56 |
+
"static_position_embeddings": true,
|
| 57 |
+
"torch_dtype": "float32",
|
| 58 |
+
"transformers_version": "4.40.2",
|
| 59 |
+
"use_cache": true,
|
| 60 |
+
"vocab_size": 64172
|
| 61 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bad_words_ids": [
|
| 3 |
+
[
|
| 4 |
+
64171
|
| 5 |
+
]
|
| 6 |
+
],
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"decoder_start_token_id": 64171,
|
| 9 |
+
"eos_token_id": 0,
|
| 10 |
+
"forced_eos_token_id": 0,
|
| 11 |
+
"max_length": 512,
|
| 12 |
+
"num_beams": 6,
|
| 13 |
+
"pad_token_id": 64171,
|
| 14 |
+
"renormalize_logits": true,
|
| 15 |
+
"transformers_version": "4.40.2"
|
| 16 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:19586fea4c6d2b80f211391f7cfe96deb3b61081461e35724eae6e6adbee055b
|
| 3 |
+
size 308263984
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1971dd810cb33487782edef3b126d43aeb420b347a80d1528acb5d0e061f06c3
|
| 3 |
+
size 616168058
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c45f065fcf14e2549693e34a722ba67b8683bc89c10fc34fe090e45f1e626b50
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0c81adbf66819e1e7e076786f12a779a7f82a200e63a4c86dbcc221ccf9b2841
|
| 3 |
+
size 1064
|
source.spm
ADDED
|
Binary file (707 kB). View file
|
|
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eos_token": {
|
| 3 |
+
"content": "</s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"pad_token": {
|
| 10 |
+
"content": "<pad>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"unk_token": {
|
| 17 |
+
"content": "<unk>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
}
|
| 23 |
+
}
|
target.spm
ADDED
|
Binary file (791 kB). View file
|
|
|
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 |
+
"64171": {
|
| 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": true,
|
| 29 |
+
"eos_token": "</s>",
|
| 30 |
+
"model_max_length": 512,
|
| 31 |
+
"pad_token": "<pad>",
|
| 32 |
+
"return_tensors": "pt",
|
| 33 |
+
"separate_vocabs": false,
|
| 34 |
+
"source_lang": "mul",
|
| 35 |
+
"sp_model_kwargs": {},
|
| 36 |
+
"target_lang": "eng",
|
| 37 |
+
"tokenizer_class": "MarianTokenizer",
|
| 38 |
+
"unk_token": "<unk>"
|
| 39 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,630 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 280.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 43960,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 3.1847133757961785,
|
| 13 |
+
"grad_norm": 3.422844171524048,
|
| 14 |
+
"learning_rate": 1.9872611464968155e-05,
|
| 15 |
+
"loss": 0.9766,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 6.369426751592357,
|
| 20 |
+
"grad_norm": 4.270854949951172,
|
| 21 |
+
"learning_rate": 1.9745222929936306e-05,
|
| 22 |
+
"loss": 0.1036,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 9.554140127388536,
|
| 27 |
+
"grad_norm": 1.767094373703003,
|
| 28 |
+
"learning_rate": 1.961783439490446e-05,
|
| 29 |
+
"loss": 0.0667,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 12.738853503184714,
|
| 34 |
+
"grad_norm": 3.3422610759735107,
|
| 35 |
+
"learning_rate": 1.9490445859872614e-05,
|
| 36 |
+
"loss": 0.0507,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 15.923566878980893,
|
| 41 |
+
"grad_norm": 2.3477580547332764,
|
| 42 |
+
"learning_rate": 1.9363057324840767e-05,
|
| 43 |
+
"loss": 0.0369,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 19.10828025477707,
|
| 48 |
+
"grad_norm": 1.2746729850769043,
|
| 49 |
+
"learning_rate": 1.9235668789808918e-05,
|
| 50 |
+
"loss": 0.0279,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 22.29299363057325,
|
| 55 |
+
"grad_norm": 1.4915257692337036,
|
| 56 |
+
"learning_rate": 1.910828025477707e-05,
|
| 57 |
+
"loss": 0.0208,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 25.477707006369428,
|
| 62 |
+
"grad_norm": 1.313347339630127,
|
| 63 |
+
"learning_rate": 1.8980891719745225e-05,
|
| 64 |
+
"loss": 0.0162,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 28.662420382165607,
|
| 69 |
+
"grad_norm": 1.810874581336975,
|
| 70 |
+
"learning_rate": 1.8853503184713376e-05,
|
| 71 |
+
"loss": 0.0124,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 31.847133757961785,
|
| 76 |
+
"grad_norm": 1.6056509017944336,
|
| 77 |
+
"learning_rate": 1.872611464968153e-05,
|
| 78 |
+
"loss": 0.0102,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 35.031847133757964,
|
| 83 |
+
"grad_norm": 1.908366084098816,
|
| 84 |
+
"learning_rate": 1.8598726114649684e-05,
|
| 85 |
+
"loss": 0.0082,
|
| 86 |
+
"step": 5500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 38.21656050955414,
|
| 90 |
+
"grad_norm": 0.49258118867874146,
|
| 91 |
+
"learning_rate": 1.8471337579617837e-05,
|
| 92 |
+
"loss": 0.0068,
|
| 93 |
+
"step": 6000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 41.40127388535032,
|
| 97 |
+
"grad_norm": 1.2429087162017822,
|
| 98 |
+
"learning_rate": 1.8343949044585988e-05,
|
| 99 |
+
"loss": 0.0066,
|
| 100 |
+
"step": 6500
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 44.5859872611465,
|
| 104 |
+
"grad_norm": 1.1771211624145508,
|
| 105 |
+
"learning_rate": 1.821656050955414e-05,
|
| 106 |
+
"loss": 0.0059,
|
| 107 |
+
"step": 7000
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 47.77070063694268,
|
| 111 |
+
"grad_norm": 0.8416564464569092,
|
| 112 |
+
"learning_rate": 1.8089171974522295e-05,
|
| 113 |
+
"loss": 0.0047,
|
| 114 |
+
"step": 7500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 50.955414012738856,
|
| 118 |
+
"grad_norm": 1.357130765914917,
|
| 119 |
+
"learning_rate": 1.796178343949045e-05,
|
| 120 |
+
"loss": 0.0044,
|
| 121 |
+
"step": 8000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 54.140127388535035,
|
| 125 |
+
"grad_norm": 0.038406338542699814,
|
| 126 |
+
"learning_rate": 1.78343949044586e-05,
|
| 127 |
+
"loss": 0.0047,
|
| 128 |
+
"step": 8500
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 57.32484076433121,
|
| 132 |
+
"grad_norm": 0.9533712267875671,
|
| 133 |
+
"learning_rate": 1.7707006369426754e-05,
|
| 134 |
+
"loss": 0.0042,
|
| 135 |
+
"step": 9000
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 60.50955414012739,
|
| 139 |
+
"grad_norm": 1.7286570072174072,
|
| 140 |
+
"learning_rate": 1.7579617834394907e-05,
|
| 141 |
+
"loss": 0.0038,
|
| 142 |
+
"step": 9500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 63.69426751592356,
|
| 146 |
+
"grad_norm": 0.576912522315979,
|
| 147 |
+
"learning_rate": 1.7452229299363058e-05,
|
| 148 |
+
"loss": 0.0037,
|
| 149 |
+
"step": 10000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 66.87898089171975,
|
| 153 |
+
"grad_norm": 0.5280445218086243,
|
| 154 |
+
"learning_rate": 1.732484076433121e-05,
|
| 155 |
+
"loss": 0.0036,
|
| 156 |
+
"step": 10500
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 70.06369426751593,
|
| 160 |
+
"grad_norm": 0.664702296257019,
|
| 161 |
+
"learning_rate": 1.7197452229299365e-05,
|
| 162 |
+
"loss": 0.0029,
|
| 163 |
+
"step": 11000
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 73.2484076433121,
|
| 167 |
+
"grad_norm": 0.18195950984954834,
|
| 168 |
+
"learning_rate": 1.707006369426752e-05,
|
| 169 |
+
"loss": 0.0033,
|
| 170 |
+
"step": 11500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 76.43312101910828,
|
| 174 |
+
"grad_norm": 0.419512003660202,
|
| 175 |
+
"learning_rate": 1.694267515923567e-05,
|
| 176 |
+
"loss": 0.0036,
|
| 177 |
+
"step": 12000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 79.61783439490446,
|
| 181 |
+
"grad_norm": 1.1403623819351196,
|
| 182 |
+
"learning_rate": 1.6815286624203824e-05,
|
| 183 |
+
"loss": 0.0033,
|
| 184 |
+
"step": 12500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 82.80254777070064,
|
| 188 |
+
"grad_norm": 1.0261355638504028,
|
| 189 |
+
"learning_rate": 1.6687898089171977e-05,
|
| 190 |
+
"loss": 0.0027,
|
| 191 |
+
"step": 13000
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 85.98726114649682,
|
| 195 |
+
"grad_norm": 0.1661386638879776,
|
| 196 |
+
"learning_rate": 1.6560509554140128e-05,
|
| 197 |
+
"loss": 0.0024,
|
| 198 |
+
"step": 13500
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 89.171974522293,
|
| 202 |
+
"grad_norm": 0.039521217346191406,
|
| 203 |
+
"learning_rate": 1.643312101910828e-05,
|
| 204 |
+
"loss": 0.0029,
|
| 205 |
+
"step": 14000
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 92.35668789808918,
|
| 209 |
+
"grad_norm": 0.047124966979026794,
|
| 210 |
+
"learning_rate": 1.6305732484076436e-05,
|
| 211 |
+
"loss": 0.0028,
|
| 212 |
+
"step": 14500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 95.54140127388536,
|
| 216 |
+
"grad_norm": 0.012884723953902721,
|
| 217 |
+
"learning_rate": 1.617834394904459e-05,
|
| 218 |
+
"loss": 0.0026,
|
| 219 |
+
"step": 15000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 98.72611464968153,
|
| 223 |
+
"grad_norm": 0.41387057304382324,
|
| 224 |
+
"learning_rate": 1.605095541401274e-05,
|
| 225 |
+
"loss": 0.0026,
|
| 226 |
+
"step": 15500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 101.91082802547771,
|
| 230 |
+
"grad_norm": 0.6253780722618103,
|
| 231 |
+
"learning_rate": 1.5923566878980894e-05,
|
| 232 |
+
"loss": 0.0025,
|
| 233 |
+
"step": 16000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 105.09554140127389,
|
| 237 |
+
"grad_norm": 1.3857741355895996,
|
| 238 |
+
"learning_rate": 1.5796178343949047e-05,
|
| 239 |
+
"loss": 0.0026,
|
| 240 |
+
"step": 16500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 108.28025477707007,
|
| 244 |
+
"grad_norm": 0.14438733458518982,
|
| 245 |
+
"learning_rate": 1.56687898089172e-05,
|
| 246 |
+
"loss": 0.0021,
|
| 247 |
+
"step": 17000
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 111.46496815286625,
|
| 251 |
+
"grad_norm": 0.0025589189026504755,
|
| 252 |
+
"learning_rate": 1.5541401273885352e-05,
|
| 253 |
+
"loss": 0.0024,
|
| 254 |
+
"step": 17500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 114.64968152866243,
|
| 258 |
+
"grad_norm": 0.008726066909730434,
|
| 259 |
+
"learning_rate": 1.5414012738853506e-05,
|
| 260 |
+
"loss": 0.002,
|
| 261 |
+
"step": 18000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 117.8343949044586,
|
| 265 |
+
"grad_norm": 0.0028913430869579315,
|
| 266 |
+
"learning_rate": 1.528662420382166e-05,
|
| 267 |
+
"loss": 0.002,
|
| 268 |
+
"step": 18500
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 121.01910828025478,
|
| 272 |
+
"grad_norm": 0.0131604615598917,
|
| 273 |
+
"learning_rate": 1.5159235668789811e-05,
|
| 274 |
+
"loss": 0.0021,
|
| 275 |
+
"step": 19000
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 124.20382165605096,
|
| 279 |
+
"grad_norm": 0.09808602929115295,
|
| 280 |
+
"learning_rate": 1.5031847133757964e-05,
|
| 281 |
+
"loss": 0.0018,
|
| 282 |
+
"step": 19500
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 127.38853503184713,
|
| 286 |
+
"grad_norm": 0.02476484701037407,
|
| 287 |
+
"learning_rate": 1.4904458598726114e-05,
|
| 288 |
+
"loss": 0.0015,
|
| 289 |
+
"step": 20000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 130.57324840764332,
|
| 293 |
+
"grad_norm": 0.015077279880642891,
|
| 294 |
+
"learning_rate": 1.4777070063694268e-05,
|
| 295 |
+
"loss": 0.0018,
|
| 296 |
+
"step": 20500
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 133.7579617834395,
|
| 300 |
+
"grad_norm": 0.06911738961935043,
|
| 301 |
+
"learning_rate": 1.464968152866242e-05,
|
| 302 |
+
"loss": 0.0022,
|
| 303 |
+
"step": 21000
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 136.94267515923568,
|
| 307 |
+
"grad_norm": 0.00874880701303482,
|
| 308 |
+
"learning_rate": 1.4522292993630574e-05,
|
| 309 |
+
"loss": 0.0019,
|
| 310 |
+
"step": 21500
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 140.12738853503186,
|
| 314 |
+
"grad_norm": 1.0872219800949097,
|
| 315 |
+
"learning_rate": 1.4394904458598726e-05,
|
| 316 |
+
"loss": 0.0015,
|
| 317 |
+
"step": 22000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 143.31210191082803,
|
| 321 |
+
"grad_norm": 0.0024208752438426018,
|
| 322 |
+
"learning_rate": 1.426751592356688e-05,
|
| 323 |
+
"loss": 0.002,
|
| 324 |
+
"step": 22500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 146.4968152866242,
|
| 328 |
+
"grad_norm": 0.0036262290086597204,
|
| 329 |
+
"learning_rate": 1.4140127388535032e-05,
|
| 330 |
+
"loss": 0.0015,
|
| 331 |
+
"step": 23000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 149.6815286624204,
|
| 335 |
+
"grad_norm": 0.03490140661597252,
|
| 336 |
+
"learning_rate": 1.4012738853503186e-05,
|
| 337 |
+
"loss": 0.0016,
|
| 338 |
+
"step": 23500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 152.86624203821657,
|
| 342 |
+
"grad_norm": 0.7626912593841553,
|
| 343 |
+
"learning_rate": 1.3885350318471338e-05,
|
| 344 |
+
"loss": 0.0015,
|
| 345 |
+
"step": 24000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 156.05095541401275,
|
| 349 |
+
"grad_norm": 0.014237442053854465,
|
| 350 |
+
"learning_rate": 1.375796178343949e-05,
|
| 351 |
+
"loss": 0.0017,
|
| 352 |
+
"step": 24500
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 159.23566878980893,
|
| 356 |
+
"grad_norm": 0.005636280868202448,
|
| 357 |
+
"learning_rate": 1.3630573248407644e-05,
|
| 358 |
+
"loss": 0.0016,
|
| 359 |
+
"step": 25000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 162.4203821656051,
|
| 363 |
+
"grad_norm": 0.05069362744688988,
|
| 364 |
+
"learning_rate": 1.3503184713375796e-05,
|
| 365 |
+
"loss": 0.0015,
|
| 366 |
+
"step": 25500
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 165.60509554140128,
|
| 370 |
+
"grad_norm": 0.0035111624747514725,
|
| 371 |
+
"learning_rate": 1.337579617834395e-05,
|
| 372 |
+
"loss": 0.0016,
|
| 373 |
+
"step": 26000
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 168.78980891719746,
|
| 377 |
+
"grad_norm": 0.002414689166471362,
|
| 378 |
+
"learning_rate": 1.3248407643312102e-05,
|
| 379 |
+
"loss": 0.0014,
|
| 380 |
+
"step": 26500
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 171.97452229299364,
|
| 384 |
+
"grad_norm": 0.05457231029868126,
|
| 385 |
+
"learning_rate": 1.3121019108280256e-05,
|
| 386 |
+
"loss": 0.0014,
|
| 387 |
+
"step": 27000
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 175.15923566878982,
|
| 391 |
+
"grad_norm": 0.0018200245685875416,
|
| 392 |
+
"learning_rate": 1.2993630573248408e-05,
|
| 393 |
+
"loss": 0.0013,
|
| 394 |
+
"step": 27500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 178.343949044586,
|
| 398 |
+
"grad_norm": 0.37822210788726807,
|
| 399 |
+
"learning_rate": 1.2866242038216562e-05,
|
| 400 |
+
"loss": 0.0014,
|
| 401 |
+
"step": 28000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 181.52866242038218,
|
| 405 |
+
"grad_norm": 0.009616321884095669,
|
| 406 |
+
"learning_rate": 1.2738853503184714e-05,
|
| 407 |
+
"loss": 0.0019,
|
| 408 |
+
"step": 28500
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 184.71337579617835,
|
| 412 |
+
"grad_norm": 0.20008017122745514,
|
| 413 |
+
"learning_rate": 1.2611464968152866e-05,
|
| 414 |
+
"loss": 0.0024,
|
| 415 |
+
"step": 29000
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 187.89808917197453,
|
| 419 |
+
"grad_norm": 1.0314069986343384,
|
| 420 |
+
"learning_rate": 1.248407643312102e-05,
|
| 421 |
+
"loss": 0.0011,
|
| 422 |
+
"step": 29500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 191.0828025477707,
|
| 426 |
+
"grad_norm": 0.0037857957649976015,
|
| 427 |
+
"learning_rate": 1.2356687898089172e-05,
|
| 428 |
+
"loss": 0.0015,
|
| 429 |
+
"step": 30000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 194.2675159235669,
|
| 433 |
+
"grad_norm": 0.07296785712242126,
|
| 434 |
+
"learning_rate": 1.2229299363057326e-05,
|
| 435 |
+
"loss": 0.0012,
|
| 436 |
+
"step": 30500
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 197.45222929936307,
|
| 440 |
+
"grad_norm": 0.25247231125831604,
|
| 441 |
+
"learning_rate": 1.2101910828025478e-05,
|
| 442 |
+
"loss": 0.0015,
|
| 443 |
+
"step": 31000
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 200.63694267515925,
|
| 447 |
+
"grad_norm": 0.008463304489850998,
|
| 448 |
+
"learning_rate": 1.1974522292993632e-05,
|
| 449 |
+
"loss": 0.0012,
|
| 450 |
+
"step": 31500
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 203.82165605095543,
|
| 454 |
+
"grad_norm": 0.0013085840037092566,
|
| 455 |
+
"learning_rate": 1.1847133757961784e-05,
|
| 456 |
+
"loss": 0.0014,
|
| 457 |
+
"step": 32000
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 207.0063694267516,
|
| 461 |
+
"grad_norm": 0.00894926954060793,
|
| 462 |
+
"learning_rate": 1.1719745222929938e-05,
|
| 463 |
+
"loss": 0.0011,
|
| 464 |
+
"step": 32500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 210.19108280254778,
|
| 468 |
+
"grad_norm": 0.016246678307652473,
|
| 469 |
+
"learning_rate": 1.159235668789809e-05,
|
| 470 |
+
"loss": 0.0011,
|
| 471 |
+
"step": 33000
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 213.37579617834396,
|
| 475 |
+
"grad_norm": 0.00870199128985405,
|
| 476 |
+
"learning_rate": 1.1464968152866242e-05,
|
| 477 |
+
"loss": 0.0011,
|
| 478 |
+
"step": 33500
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 216.56050955414014,
|
| 482 |
+
"grad_norm": 0.3081636130809784,
|
| 483 |
+
"learning_rate": 1.1337579617834396e-05,
|
| 484 |
+
"loss": 0.001,
|
| 485 |
+
"step": 34000
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 219.74522292993632,
|
| 489 |
+
"grad_norm": 0.7328362464904785,
|
| 490 |
+
"learning_rate": 1.1210191082802548e-05,
|
| 491 |
+
"loss": 0.0016,
|
| 492 |
+
"step": 34500
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 222.9299363057325,
|
| 496 |
+
"grad_norm": 0.0006987511878833175,
|
| 497 |
+
"learning_rate": 1.1082802547770702e-05,
|
| 498 |
+
"loss": 0.0012,
|
| 499 |
+
"step": 35000
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 226.11464968152868,
|
| 503 |
+
"grad_norm": 0.016234412789344788,
|
| 504 |
+
"learning_rate": 1.0955414012738854e-05,
|
| 505 |
+
"loss": 0.0011,
|
| 506 |
+
"step": 35500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 229.29936305732485,
|
| 510 |
+
"grad_norm": 0.7393547892570496,
|
| 511 |
+
"learning_rate": 1.0828025477707008e-05,
|
| 512 |
+
"loss": 0.0011,
|
| 513 |
+
"step": 36000
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 232.48407643312103,
|
| 517 |
+
"grad_norm": 0.0006552963168360293,
|
| 518 |
+
"learning_rate": 1.070063694267516e-05,
|
| 519 |
+
"loss": 0.0009,
|
| 520 |
+
"step": 36500
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 235.6687898089172,
|
| 524 |
+
"grad_norm": 0.003285791492089629,
|
| 525 |
+
"learning_rate": 1.0573248407643314e-05,
|
| 526 |
+
"loss": 0.0011,
|
| 527 |
+
"step": 37000
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 238.8535031847134,
|
| 531 |
+
"grad_norm": 0.8827471733093262,
|
| 532 |
+
"learning_rate": 1.0445859872611466e-05,
|
| 533 |
+
"loss": 0.0009,
|
| 534 |
+
"step": 37500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 242.03821656050957,
|
| 538 |
+
"grad_norm": 0.03062291443347931,
|
| 539 |
+
"learning_rate": 1.0318471337579618e-05,
|
| 540 |
+
"loss": 0.001,
|
| 541 |
+
"step": 38000
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 245.22292993630575,
|
| 545 |
+
"grad_norm": 1.2000418901443481,
|
| 546 |
+
"learning_rate": 1.0191082802547772e-05,
|
| 547 |
+
"loss": 0.0014,
|
| 548 |
+
"step": 38500
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 248.40764331210192,
|
| 552 |
+
"grad_norm": 0.5093538761138916,
|
| 553 |
+
"learning_rate": 1.0063694267515924e-05,
|
| 554 |
+
"loss": 0.0009,
|
| 555 |
+
"step": 39000
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 251.59235668789808,
|
| 559 |
+
"grad_norm": 0.005525712389498949,
|
| 560 |
+
"learning_rate": 9.936305732484078e-06,
|
| 561 |
+
"loss": 0.0008,
|
| 562 |
+
"step": 39500
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 254.77707006369425,
|
| 566 |
+
"grad_norm": 1.1260795593261719,
|
| 567 |
+
"learning_rate": 9.80891719745223e-06,
|
| 568 |
+
"loss": 0.0011,
|
| 569 |
+
"step": 40000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 257.96178343949043,
|
| 573 |
+
"grad_norm": 0.0011535738594830036,
|
| 574 |
+
"learning_rate": 9.681528662420384e-06,
|
| 575 |
+
"loss": 0.0012,
|
| 576 |
+
"step": 40500
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 261.14649681528664,
|
| 580 |
+
"grad_norm": 0.009397489950060844,
|
| 581 |
+
"learning_rate": 9.554140127388536e-06,
|
| 582 |
+
"loss": 0.0012,
|
| 583 |
+
"step": 41000
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 264.3312101910828,
|
| 587 |
+
"grad_norm": 0.003204810433089733,
|
| 588 |
+
"learning_rate": 9.426751592356688e-06,
|
| 589 |
+
"loss": 0.0008,
|
| 590 |
+
"step": 41500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 267.515923566879,
|
| 594 |
+
"grad_norm": 0.0012358404928818345,
|
| 595 |
+
"learning_rate": 9.299363057324842e-06,
|
| 596 |
+
"loss": 0.0009,
|
| 597 |
+
"step": 42000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 270.70063694267515,
|
| 601 |
+
"grad_norm": 0.006686298176646233,
|
| 602 |
+
"learning_rate": 9.171974522292994e-06,
|
| 603 |
+
"loss": 0.0011,
|
| 604 |
+
"step": 42500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 273.88535031847135,
|
| 608 |
+
"grad_norm": 0.004616052843630314,
|
| 609 |
+
"learning_rate": 9.044585987261148e-06,
|
| 610 |
+
"loss": 0.001,
|
| 611 |
+
"step": 43000
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 277.0700636942675,
|
| 615 |
+
"grad_norm": 0.000694330723490566,
|
| 616 |
+
"learning_rate": 8.9171974522293e-06,
|
| 617 |
+
"loss": 0.0009,
|
| 618 |
+
"step": 43500
|
| 619 |
+
}
|
| 620 |
+
],
|
| 621 |
+
"logging_steps": 500,
|
| 622 |
+
"max_steps": 78500,
|
| 623 |
+
"num_input_tokens_seen": 0,
|
| 624 |
+
"num_train_epochs": 500,
|
| 625 |
+
"save_steps": 500,
|
| 626 |
+
"total_flos": 9164086093283328.0,
|
| 627 |
+
"train_batch_size": 32,
|
| 628 |
+
"trial_name": null,
|
| 629 |
+
"trial_params": null
|
| 630 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:37b7e9dcf130a0c92d1859cc8d5157e549bf8fbc5baaff814314da6030b046b7
|
| 3 |
+
size 5432
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|