Upload 13 files
Browse files- .gitattributes +2 -0
- all_results.json +16 -0
- config.json +56 -0
- eval_results.json +9 -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
- train_results.json +11 -0
- trainer_state.json +1865 -0
- training_args.bin +3 -0
- vocab.json +0 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,5 @@ 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 |
+
source.spm filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
target.spm filter=lfs diff=lfs merge=lfs -text
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 2.435030460357666,
|
| 4 |
+
"eval_runtime": 2.5197,
|
| 5 |
+
"eval_samples": 2500,
|
| 6 |
+
"eval_samples_per_second": 992.173,
|
| 7 |
+
"eval_steps_per_second": 124.22,
|
| 8 |
+
"num_input_tokens_seen": 55653732,
|
| 9 |
+
"total_flos": 1.4738832163602432e+16,
|
| 10 |
+
"train_loss": 1.7073542784139526,
|
| 11 |
+
"train_runtime": 2504.3889,
|
| 12 |
+
"train_samples": 248732,
|
| 13 |
+
"train_samples_per_second": 297.955,
|
| 14 |
+
"train_steps_per_second": 37.245,
|
| 15 |
+
"train_tokens_per_second": 22216.164
|
| 16 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Helsinki-NLP/opus-mt-zh-en",
|
| 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 |
+
"bos_token_id": 0,
|
| 12 |
+
"classif_dropout": 0.0,
|
| 13 |
+
"classifier_dropout": 0.0,
|
| 14 |
+
"d_model": 512,
|
| 15 |
+
"decoder_attention_heads": 8,
|
| 16 |
+
"decoder_ffn_dim": 2048,
|
| 17 |
+
"decoder_layerdrop": 0.0,
|
| 18 |
+
"decoder_layers": 6,
|
| 19 |
+
"decoder_start_token_id": 65000,
|
| 20 |
+
"decoder_vocab_size": 65001,
|
| 21 |
+
"dropout": 0.1,
|
| 22 |
+
"encoder_attention_heads": 8,
|
| 23 |
+
"encoder_ffn_dim": 2048,
|
| 24 |
+
"encoder_layerdrop": 0.0,
|
| 25 |
+
"encoder_layers": 6,
|
| 26 |
+
"eos_token_id": 0,
|
| 27 |
+
"extra_pos_embeddings": 65001,
|
| 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": 65000,
|
| 49 |
+
"scale_embedding": true,
|
| 50 |
+
"share_encoder_decoder_embeddings": true,
|
| 51 |
+
"static_position_embeddings": true,
|
| 52 |
+
"torch_dtype": "float32",
|
| 53 |
+
"transformers_version": "4.48.1",
|
| 54 |
+
"use_cache": true,
|
| 55 |
+
"vocab_size": 65001
|
| 56 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 2.435030460357666,
|
| 4 |
+
"eval_runtime": 2.5197,
|
| 5 |
+
"eval_samples": 2500,
|
| 6 |
+
"eval_samples_per_second": 992.173,
|
| 7 |
+
"eval_steps_per_second": 124.22,
|
| 8 |
+
"num_input_tokens_seen": 55653732
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bad_words_ids": [
|
| 3 |
+
[
|
| 4 |
+
65000
|
| 5 |
+
]
|
| 6 |
+
],
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"decoder_start_token_id": 65000,
|
| 9 |
+
"eos_token_id": 0,
|
| 10 |
+
"forced_eos_token_id": 0,
|
| 11 |
+
"max_length": 512,
|
| 12 |
+
"num_beams": 6,
|
| 13 |
+
"pad_token_id": 65000,
|
| 14 |
+
"renormalize_logits": true,
|
| 15 |
+
"transformers_version": "4.48.1"
|
| 16 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7f894a6d080dfe10bed879a7505ab686823e3d5ba15eed0f44ea70a17462fe4a
|
| 3 |
+
size 309965092
|
source.spm
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e27a3a1b539f4959ec72ea60e453f49156289f95d4e6000b29332efc45616203
|
| 3 |
+
size 804677
|
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 sha256:6a881f4717cd7265f53fea54fd3dc689c767c05338fac7a4590f3088cb2d7855
|
| 3 |
+
size 806530
|
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 |
+
"65000": {
|
| 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": "zho",
|
| 35 |
+
"sp_model_kwargs": {},
|
| 36 |
+
"target_lang": "eng",
|
| 37 |
+
"tokenizer_class": "MarianTokenizer",
|
| 38 |
+
"unk_token": "<unk>"
|
| 39 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"num_input_tokens_seen": 55653732,
|
| 4 |
+
"total_flos": 1.4738832163602432e+16,
|
| 5 |
+
"train_loss": 1.7073542784139526,
|
| 6 |
+
"train_runtime": 2504.3889,
|
| 7 |
+
"train_samples": 248732,
|
| 8 |
+
"train_samples_per_second": 297.955,
|
| 9 |
+
"train_steps_per_second": 37.245,
|
| 10 |
+
"train_tokens_per_second": 22216.164
|
| 11 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1865 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 2.435030460357666,
|
| 3 |
+
"best_model_checkpoint": "zhtw-en/checkpoint-92500",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 2500,
|
| 6 |
+
"global_step": 93276,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.016081307088640164,
|
| 13 |
+
"grad_norm": 8.550692558288574,
|
| 14 |
+
"learning_rate": 4.919593464556799e-05,
|
| 15 |
+
"loss": 3.4707,
|
| 16 |
+
"num_input_tokens_seen": 301408,
|
| 17 |
+
"step": 500
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.03216261417728033,
|
| 21 |
+
"grad_norm": 9.999076843261719,
|
| 22 |
+
"learning_rate": 4.8391869291135987e-05,
|
| 23 |
+
"loss": 3.3664,
|
| 24 |
+
"num_input_tokens_seen": 597672,
|
| 25 |
+
"step": 1000
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.048243921265920496,
|
| 29 |
+
"grad_norm": 8.839406967163086,
|
| 30 |
+
"learning_rate": 4.7587803936703975e-05,
|
| 31 |
+
"loss": 3.3089,
|
| 32 |
+
"num_input_tokens_seen": 896520,
|
| 33 |
+
"step": 1500
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.06432522835456066,
|
| 37 |
+
"grad_norm": 7.105090618133545,
|
| 38 |
+
"learning_rate": 4.678373858227197e-05,
|
| 39 |
+
"loss": 3.267,
|
| 40 |
+
"num_input_tokens_seen": 1194832,
|
| 41 |
+
"step": 2000
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.08040653544320082,
|
| 45 |
+
"grad_norm": 7.992733001708984,
|
| 46 |
+
"learning_rate": 4.597967322783996e-05,
|
| 47 |
+
"loss": 3.2254,
|
| 48 |
+
"num_input_tokens_seen": 1493088,
|
| 49 |
+
"step": 2500
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.08040653544320082,
|
| 53 |
+
"eval_loss": 2.910461664199829,
|
| 54 |
+
"eval_runtime": 2.5005,
|
| 55 |
+
"eval_samples_per_second": 999.788,
|
| 56 |
+
"eval_steps_per_second": 125.173,
|
| 57 |
+
"num_input_tokens_seen": 1493088,
|
| 58 |
+
"step": 2500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.09648784253184099,
|
| 62 |
+
"grad_norm": 8.68657112121582,
|
| 63 |
+
"learning_rate": 4.5175607873407955e-05,
|
| 64 |
+
"loss": 3.1691,
|
| 65 |
+
"num_input_tokens_seen": 1793976,
|
| 66 |
+
"step": 3000
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.11256914962048115,
|
| 70 |
+
"grad_norm": 7.246800899505615,
|
| 71 |
+
"learning_rate": 4.4371542518975943e-05,
|
| 72 |
+
"loss": 3.1685,
|
| 73 |
+
"num_input_tokens_seen": 2095352,
|
| 74 |
+
"step": 3500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.1286504567091213,
|
| 78 |
+
"grad_norm": 9.030860900878906,
|
| 79 |
+
"learning_rate": 4.356747716454393e-05,
|
| 80 |
+
"loss": 3.1333,
|
| 81 |
+
"num_input_tokens_seen": 2393856,
|
| 82 |
+
"step": 4000
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.14473176379776148,
|
| 86 |
+
"grad_norm": 7.463845252990723,
|
| 87 |
+
"learning_rate": 4.276341181011193e-05,
|
| 88 |
+
"loss": 3.1295,
|
| 89 |
+
"num_input_tokens_seen": 2694496,
|
| 90 |
+
"step": 4500
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.16081307088640165,
|
| 94 |
+
"grad_norm": 8.482089042663574,
|
| 95 |
+
"learning_rate": 4.195934645567992e-05,
|
| 96 |
+
"loss": 3.0946,
|
| 97 |
+
"num_input_tokens_seen": 2990968,
|
| 98 |
+
"step": 5000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.16081307088640165,
|
| 102 |
+
"eval_loss": 2.830476999282837,
|
| 103 |
+
"eval_runtime": 2.5201,
|
| 104 |
+
"eval_samples_per_second": 992.009,
|
| 105 |
+
"eval_steps_per_second": 124.199,
|
| 106 |
+
"num_input_tokens_seen": 2990968,
|
| 107 |
+
"step": 5000
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.17689437797504182,
|
| 111 |
+
"grad_norm": 8.570518493652344,
|
| 112 |
+
"learning_rate": 4.115528110124791e-05,
|
| 113 |
+
"loss": 3.112,
|
| 114 |
+
"num_input_tokens_seen": 3289488,
|
| 115 |
+
"step": 5500
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.19297568506368198,
|
| 119 |
+
"grad_norm": 9.759325981140137,
|
| 120 |
+
"learning_rate": 4.03512157468159e-05,
|
| 121 |
+
"loss": 3.0933,
|
| 122 |
+
"num_input_tokens_seen": 3590264,
|
| 123 |
+
"step": 6000
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.20905699215232215,
|
| 127 |
+
"grad_norm": 6.518988609313965,
|
| 128 |
+
"learning_rate": 3.9547150392383896e-05,
|
| 129 |
+
"loss": 3.0858,
|
| 130 |
+
"num_input_tokens_seen": 3885160,
|
| 131 |
+
"step": 6500
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.2251382992409623,
|
| 135 |
+
"grad_norm": 6.913475036621094,
|
| 136 |
+
"learning_rate": 3.8743085037951885e-05,
|
| 137 |
+
"loss": 3.0543,
|
| 138 |
+
"num_input_tokens_seen": 4184600,
|
| 139 |
+
"step": 7000
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.24121960632960246,
|
| 143 |
+
"grad_norm": 8.485562324523926,
|
| 144 |
+
"learning_rate": 3.793901968351988e-05,
|
| 145 |
+
"loss": 3.0473,
|
| 146 |
+
"num_input_tokens_seen": 4477792,
|
| 147 |
+
"step": 7500
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.24121960632960246,
|
| 151 |
+
"eval_loss": 2.773728847503662,
|
| 152 |
+
"eval_runtime": 2.5738,
|
| 153 |
+
"eval_samples_per_second": 971.323,
|
| 154 |
+
"eval_steps_per_second": 121.61,
|
| 155 |
+
"num_input_tokens_seen": 4477792,
|
| 156 |
+
"step": 7500
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.2573009134182426,
|
| 160 |
+
"grad_norm": 7.89262056350708,
|
| 161 |
+
"learning_rate": 3.713495432908787e-05,
|
| 162 |
+
"loss": 3.029,
|
| 163 |
+
"num_input_tokens_seen": 4779520,
|
| 164 |
+
"step": 8000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.2733822205068828,
|
| 168 |
+
"grad_norm": 6.879751205444336,
|
| 169 |
+
"learning_rate": 3.6330888974655864e-05,
|
| 170 |
+
"loss": 3.0127,
|
| 171 |
+
"num_input_tokens_seen": 5078952,
|
| 172 |
+
"step": 8500
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.28946352759552296,
|
| 176 |
+
"grad_norm": 8.109273910522461,
|
| 177 |
+
"learning_rate": 3.552682362022385e-05,
|
| 178 |
+
"loss": 3.0078,
|
| 179 |
+
"num_input_tokens_seen": 5376128,
|
| 180 |
+
"step": 9000
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 0.3055448346841631,
|
| 184 |
+
"grad_norm": 8.074146270751953,
|
| 185 |
+
"learning_rate": 3.472275826579184e-05,
|
| 186 |
+
"loss": 2.9988,
|
| 187 |
+
"num_input_tokens_seen": 5671664,
|
| 188 |
+
"step": 9500
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.3216261417728033,
|
| 192 |
+
"grad_norm": 6.523529529571533,
|
| 193 |
+
"learning_rate": 3.391869291135984e-05,
|
| 194 |
+
"loss": 2.9633,
|
| 195 |
+
"num_input_tokens_seen": 5967560,
|
| 196 |
+
"step": 10000
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.3216261417728033,
|
| 200 |
+
"eval_loss": 2.7306864261627197,
|
| 201 |
+
"eval_runtime": 2.505,
|
| 202 |
+
"eval_samples_per_second": 997.986,
|
| 203 |
+
"eval_steps_per_second": 124.948,
|
| 204 |
+
"num_input_tokens_seen": 5967560,
|
| 205 |
+
"step": 10000
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.33770744886144344,
|
| 209 |
+
"grad_norm": 6.974269866943359,
|
| 210 |
+
"learning_rate": 3.311462755692783e-05,
|
| 211 |
+
"loss": 2.9732,
|
| 212 |
+
"num_input_tokens_seen": 6265312,
|
| 213 |
+
"step": 10500
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.35378875595008363,
|
| 217 |
+
"grad_norm": 7.644798278808594,
|
| 218 |
+
"learning_rate": 3.231056220249582e-05,
|
| 219 |
+
"loss": 2.9729,
|
| 220 |
+
"num_input_tokens_seen": 6563632,
|
| 221 |
+
"step": 11000
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.3698700630387238,
|
| 225 |
+
"grad_norm": 7.96437406539917,
|
| 226 |
+
"learning_rate": 3.150649684806381e-05,
|
| 227 |
+
"loss": 2.9484,
|
| 228 |
+
"num_input_tokens_seen": 6865528,
|
| 229 |
+
"step": 11500
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 0.38595137012736397,
|
| 233 |
+
"grad_norm": 7.939519882202148,
|
| 234 |
+
"learning_rate": 3.0702431493631805e-05,
|
| 235 |
+
"loss": 2.9387,
|
| 236 |
+
"num_input_tokens_seen": 7165632,
|
| 237 |
+
"step": 12000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.4020326772160041,
|
| 241 |
+
"grad_norm": 7.698306083679199,
|
| 242 |
+
"learning_rate": 2.9898366139199797e-05,
|
| 243 |
+
"loss": 2.9355,
|
| 244 |
+
"num_input_tokens_seen": 7463192,
|
| 245 |
+
"step": 12500
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.4020326772160041,
|
| 249 |
+
"eval_loss": 2.684298515319824,
|
| 250 |
+
"eval_runtime": 2.5214,
|
| 251 |
+
"eval_samples_per_second": 991.52,
|
| 252 |
+
"eval_steps_per_second": 124.138,
|
| 253 |
+
"num_input_tokens_seen": 7463192,
|
| 254 |
+
"step": 12500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.4181139843046443,
|
| 258 |
+
"grad_norm": 7.21583890914917,
|
| 259 |
+
"learning_rate": 2.9094300784767786e-05,
|
| 260 |
+
"loss": 2.9418,
|
| 261 |
+
"num_input_tokens_seen": 7758024,
|
| 262 |
+
"step": 13000
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.43419529139328444,
|
| 266 |
+
"grad_norm": 7.767180919647217,
|
| 267 |
+
"learning_rate": 2.8290235430335778e-05,
|
| 268 |
+
"loss": 2.923,
|
| 269 |
+
"num_input_tokens_seen": 8052032,
|
| 270 |
+
"step": 13500
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 0.4502765984819246,
|
| 274 |
+
"grad_norm": 7.057159423828125,
|
| 275 |
+
"learning_rate": 2.7486170075903773e-05,
|
| 276 |
+
"loss": 2.9016,
|
| 277 |
+
"num_input_tokens_seen": 8347768,
|
| 278 |
+
"step": 14000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.4663579055705648,
|
| 282 |
+
"grad_norm": 7.320003032684326,
|
| 283 |
+
"learning_rate": 2.6682104721471762e-05,
|
| 284 |
+
"loss": 2.885,
|
| 285 |
+
"num_input_tokens_seen": 8646192,
|
| 286 |
+
"step": 14500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.4824392126592049,
|
| 290 |
+
"grad_norm": 7.630561828613281,
|
| 291 |
+
"learning_rate": 2.587803936703975e-05,
|
| 292 |
+
"loss": 2.9076,
|
| 293 |
+
"num_input_tokens_seen": 8950264,
|
| 294 |
+
"step": 15000
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.4824392126592049,
|
| 298 |
+
"eval_loss": 2.658709764480591,
|
| 299 |
+
"eval_runtime": 2.6566,
|
| 300 |
+
"eval_samples_per_second": 941.035,
|
| 301 |
+
"eval_steps_per_second": 117.818,
|
| 302 |
+
"num_input_tokens_seen": 8950264,
|
| 303 |
+
"step": 15000
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.4985205197478451,
|
| 307 |
+
"grad_norm": 6.413602828979492,
|
| 308 |
+
"learning_rate": 2.507397401260775e-05,
|
| 309 |
+
"loss": 2.8818,
|
| 310 |
+
"num_input_tokens_seen": 9247272,
|
| 311 |
+
"step": 15500
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.5146018268364853,
|
| 315 |
+
"grad_norm": 6.833747863769531,
|
| 316 |
+
"learning_rate": 2.426990865817574e-05,
|
| 317 |
+
"loss": 2.8998,
|
| 318 |
+
"num_input_tokens_seen": 9547432,
|
| 319 |
+
"step": 16000
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 0.5306831339251254,
|
| 323 |
+
"grad_norm": 7.930506706237793,
|
| 324 |
+
"learning_rate": 2.3465843303743727e-05,
|
| 325 |
+
"loss": 2.8543,
|
| 326 |
+
"num_input_tokens_seen": 9844072,
|
| 327 |
+
"step": 16500
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 0.5467644410137656,
|
| 331 |
+
"grad_norm": 6.841344356536865,
|
| 332 |
+
"learning_rate": 2.2661777949311722e-05,
|
| 333 |
+
"loss": 2.8669,
|
| 334 |
+
"num_input_tokens_seen": 10142344,
|
| 335 |
+
"step": 17000
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 0.5628457481024057,
|
| 339 |
+
"grad_norm": 6.899343967437744,
|
| 340 |
+
"learning_rate": 2.185771259487971e-05,
|
| 341 |
+
"loss": 2.8714,
|
| 342 |
+
"num_input_tokens_seen": 10443344,
|
| 343 |
+
"step": 17500
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 0.5628457481024057,
|
| 347 |
+
"eval_loss": 2.6303601264953613,
|
| 348 |
+
"eval_runtime": 2.5354,
|
| 349 |
+
"eval_samples_per_second": 986.046,
|
| 350 |
+
"eval_steps_per_second": 123.453,
|
| 351 |
+
"num_input_tokens_seen": 10443344,
|
| 352 |
+
"step": 17500
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.5789270551910459,
|
| 356 |
+
"grad_norm": 6.256689071655273,
|
| 357 |
+
"learning_rate": 2.1053647240447703e-05,
|
| 358 |
+
"loss": 2.8418,
|
| 359 |
+
"num_input_tokens_seen": 10744312,
|
| 360 |
+
"step": 18000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.5950083622796861,
|
| 364 |
+
"grad_norm": 7.627821445465088,
|
| 365 |
+
"learning_rate": 2.0249581886015695e-05,
|
| 366 |
+
"loss": 2.8462,
|
| 367 |
+
"num_input_tokens_seen": 11048208,
|
| 368 |
+
"step": 18500
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.6110896693683262,
|
| 372 |
+
"grad_norm": 7.331953525543213,
|
| 373 |
+
"learning_rate": 1.9445516531583687e-05,
|
| 374 |
+
"loss": 2.8345,
|
| 375 |
+
"num_input_tokens_seen": 11347880,
|
| 376 |
+
"step": 19000
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.6271709764569664,
|
| 380 |
+
"grad_norm": 6.463207244873047,
|
| 381 |
+
"learning_rate": 1.864145117715168e-05,
|
| 382 |
+
"loss": 2.8531,
|
| 383 |
+
"num_input_tokens_seen": 11650144,
|
| 384 |
+
"step": 19500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.6432522835456066,
|
| 388 |
+
"grad_norm": 7.423746109008789,
|
| 389 |
+
"learning_rate": 1.783738582271967e-05,
|
| 390 |
+
"loss": 2.8716,
|
| 391 |
+
"num_input_tokens_seen": 11951096,
|
| 392 |
+
"step": 20000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.6432522835456066,
|
| 396 |
+
"eval_loss": 2.6024744510650635,
|
| 397 |
+
"eval_runtime": 2.5182,
|
| 398 |
+
"eval_samples_per_second": 992.755,
|
| 399 |
+
"eval_steps_per_second": 124.293,
|
| 400 |
+
"num_input_tokens_seen": 11951096,
|
| 401 |
+
"step": 20000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.6593335906342468,
|
| 405 |
+
"grad_norm": 7.352589130401611,
|
| 406 |
+
"learning_rate": 1.7033320468287664e-05,
|
| 407 |
+
"loss": 2.8243,
|
| 408 |
+
"num_input_tokens_seen": 12252592,
|
| 409 |
+
"step": 20500
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.6754148977228869,
|
| 413 |
+
"grad_norm": 7.22981071472168,
|
| 414 |
+
"learning_rate": 1.6229255113855656e-05,
|
| 415 |
+
"loss": 2.8454,
|
| 416 |
+
"num_input_tokens_seen": 12546792,
|
| 417 |
+
"step": 21000
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.6914962048115271,
|
| 421 |
+
"grad_norm": 6.819567680358887,
|
| 422 |
+
"learning_rate": 1.5425189759423648e-05,
|
| 423 |
+
"loss": 2.8047,
|
| 424 |
+
"num_input_tokens_seen": 12838728,
|
| 425 |
+
"step": 21500
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.7075775119001673,
|
| 429 |
+
"grad_norm": 8.716426849365234,
|
| 430 |
+
"learning_rate": 1.4621124404991638e-05,
|
| 431 |
+
"loss": 2.8144,
|
| 432 |
+
"num_input_tokens_seen": 13137688,
|
| 433 |
+
"step": 22000
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.7236588189888075,
|
| 437 |
+
"grad_norm": 7.324875831604004,
|
| 438 |
+
"learning_rate": 1.381705905055963e-05,
|
| 439 |
+
"loss": 2.7989,
|
| 440 |
+
"num_input_tokens_seen": 13432464,
|
| 441 |
+
"step": 22500
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.7236588189888075,
|
| 445 |
+
"eval_loss": 2.5822224617004395,
|
| 446 |
+
"eval_runtime": 2.5158,
|
| 447 |
+
"eval_samples_per_second": 993.705,
|
| 448 |
+
"eval_steps_per_second": 124.412,
|
| 449 |
+
"num_input_tokens_seen": 13432464,
|
| 450 |
+
"step": 22500
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.7397401260774475,
|
| 454 |
+
"grad_norm": 7.962778568267822,
|
| 455 |
+
"learning_rate": 1.301299369612762e-05,
|
| 456 |
+
"loss": 2.7653,
|
| 457 |
+
"num_input_tokens_seen": 13730024,
|
| 458 |
+
"step": 23000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.7558214331660877,
|
| 462 |
+
"grad_norm": 6.807019233703613,
|
| 463 |
+
"learning_rate": 1.2208928341695614e-05,
|
| 464 |
+
"loss": 2.7933,
|
| 465 |
+
"num_input_tokens_seen": 14026400,
|
| 466 |
+
"step": 23500
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.7719027402547279,
|
| 470 |
+
"grad_norm": 8.716556549072266,
|
| 471 |
+
"learning_rate": 1.1404862987263605e-05,
|
| 472 |
+
"loss": 2.7988,
|
| 473 |
+
"num_input_tokens_seen": 14326608,
|
| 474 |
+
"step": 24000
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.787984047343368,
|
| 478 |
+
"grad_norm": 7.388988018035889,
|
| 479 |
+
"learning_rate": 1.0600797632831597e-05,
|
| 480 |
+
"loss": 2.7928,
|
| 481 |
+
"num_input_tokens_seen": 14623864,
|
| 482 |
+
"step": 24500
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.8040653544320082,
|
| 486 |
+
"grad_norm": 7.011099815368652,
|
| 487 |
+
"learning_rate": 9.796732278399589e-06,
|
| 488 |
+
"loss": 2.7941,
|
| 489 |
+
"num_input_tokens_seen": 14919424,
|
| 490 |
+
"step": 25000
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 0.8040653544320082,
|
| 494 |
+
"eval_loss": 2.5630149841308594,
|
| 495 |
+
"eval_runtime": 2.5601,
|
| 496 |
+
"eval_samples_per_second": 976.534,
|
| 497 |
+
"eval_steps_per_second": 122.262,
|
| 498 |
+
"num_input_tokens_seen": 14919424,
|
| 499 |
+
"step": 25000
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.8201466615206484,
|
| 503 |
+
"grad_norm": 6.740393161773682,
|
| 504 |
+
"learning_rate": 8.992666923967581e-06,
|
| 505 |
+
"loss": 2.8089,
|
| 506 |
+
"num_input_tokens_seen": 15216136,
|
| 507 |
+
"step": 25500
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.8362279686092886,
|
| 511 |
+
"grad_norm": 7.124479293823242,
|
| 512 |
+
"learning_rate": 8.188601569535573e-06,
|
| 513 |
+
"loss": 2.7704,
|
| 514 |
+
"num_input_tokens_seen": 15515592,
|
| 515 |
+
"step": 26000
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.8523092756979287,
|
| 519 |
+
"grad_norm": 7.781102180480957,
|
| 520 |
+
"learning_rate": 7.384536215103564e-06,
|
| 521 |
+
"loss": 2.8022,
|
| 522 |
+
"num_input_tokens_seen": 15818560,
|
| 523 |
+
"step": 26500
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 0.8683905827865689,
|
| 527 |
+
"grad_norm": 6.861135005950928,
|
| 528 |
+
"learning_rate": 6.580470860671556e-06,
|
| 529 |
+
"loss": 2.7891,
|
| 530 |
+
"num_input_tokens_seen": 16114056,
|
| 531 |
+
"step": 27000
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 0.8844718898752091,
|
| 535 |
+
"grad_norm": 7.128973484039307,
|
| 536 |
+
"learning_rate": 5.776405506239547e-06,
|
| 537 |
+
"loss": 2.7692,
|
| 538 |
+
"num_input_tokens_seen": 16415080,
|
| 539 |
+
"step": 27500
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 0.8844718898752091,
|
| 543 |
+
"eval_loss": 2.5496785640716553,
|
| 544 |
+
"eval_runtime": 2.6422,
|
| 545 |
+
"eval_samples_per_second": 946.175,
|
| 546 |
+
"eval_steps_per_second": 118.461,
|
| 547 |
+
"num_input_tokens_seen": 16415080,
|
| 548 |
+
"step": 27500
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.9005531969638492,
|
| 552 |
+
"grad_norm": 8.560084342956543,
|
| 553 |
+
"learning_rate": 4.9723401518075395e-06,
|
| 554 |
+
"loss": 2.7627,
|
| 555 |
+
"num_input_tokens_seen": 16711136,
|
| 556 |
+
"step": 28000
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.9166345040524894,
|
| 560 |
+
"grad_norm": 7.5000224113464355,
|
| 561 |
+
"learning_rate": 4.168274797375531e-06,
|
| 562 |
+
"loss": 2.7687,
|
| 563 |
+
"num_input_tokens_seen": 17005880,
|
| 564 |
+
"step": 28500
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.9327158111411296,
|
| 568 |
+
"grad_norm": 6.699025630950928,
|
| 569 |
+
"learning_rate": 3.3642094429435228e-06,
|
| 570 |
+
"loss": 2.779,
|
| 571 |
+
"num_input_tokens_seen": 17307064,
|
| 572 |
+
"step": 29000
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.9487971182297698,
|
| 576 |
+
"grad_norm": 6.6417131423950195,
|
| 577 |
+
"learning_rate": 2.560144088511514e-06,
|
| 578 |
+
"loss": 2.7493,
|
| 579 |
+
"num_input_tokens_seen": 17602296,
|
| 580 |
+
"step": 29500
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 0.9648784253184098,
|
| 584 |
+
"grad_norm": 6.775792121887207,
|
| 585 |
+
"learning_rate": 1.756078734079506e-06,
|
| 586 |
+
"loss": 2.757,
|
| 587 |
+
"num_input_tokens_seen": 17897832,
|
| 588 |
+
"step": 30000
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 0.9648784253184098,
|
| 592 |
+
"eval_loss": 2.5388031005859375,
|
| 593 |
+
"eval_runtime": 2.6123,
|
| 594 |
+
"eval_samples_per_second": 957.018,
|
| 595 |
+
"eval_steps_per_second": 119.819,
|
| 596 |
+
"num_input_tokens_seen": 17897832,
|
| 597 |
+
"step": 30000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.98095973240705,
|
| 601 |
+
"grad_norm": 7.619235038757324,
|
| 602 |
+
"learning_rate": 9.520133796474978e-07,
|
| 603 |
+
"loss": 2.7433,
|
| 604 |
+
"num_input_tokens_seen": 18195568,
|
| 605 |
+
"step": 30500
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.9970410394956902,
|
| 609 |
+
"grad_norm": 6.682379722595215,
|
| 610 |
+
"learning_rate": 1.479480252154895e-07,
|
| 611 |
+
"loss": 2.7266,
|
| 612 |
+
"num_input_tokens_seen": 18491904,
|
| 613 |
+
"step": 31000
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 1.0131223465843304,
|
| 617 |
+
"grad_norm": 6.5948309898376465,
|
| 618 |
+
"learning_rate": 3.311462755692783e-05,
|
| 619 |
+
"loss": 2.6645,
|
| 620 |
+
"num_input_tokens_seen": 18790628,
|
| 621 |
+
"step": 31500
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 1.0292036536729705,
|
| 625 |
+
"grad_norm": 6.920671463012695,
|
| 626 |
+
"learning_rate": 3.2846605772117164e-05,
|
| 627 |
+
"loss": 2.6881,
|
| 628 |
+
"num_input_tokens_seen": 19090780,
|
| 629 |
+
"step": 32000
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 1.0452849607616108,
|
| 633 |
+
"grad_norm": 6.296219348907471,
|
| 634 |
+
"learning_rate": 3.257858398730649e-05,
|
| 635 |
+
"loss": 2.7024,
|
| 636 |
+
"num_input_tokens_seen": 19384812,
|
| 637 |
+
"step": 32500
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 1.0452849607616108,
|
| 641 |
+
"eval_loss": 2.6005640029907227,
|
| 642 |
+
"eval_runtime": 2.5084,
|
| 643 |
+
"eval_samples_per_second": 996.636,
|
| 644 |
+
"eval_steps_per_second": 124.779,
|
| 645 |
+
"num_input_tokens_seen": 19384812,
|
| 646 |
+
"step": 32500
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.061366267850251,
|
| 650 |
+
"grad_norm": 7.068648815155029,
|
| 651 |
+
"learning_rate": 3.231056220249582e-05,
|
| 652 |
+
"loss": 2.6939,
|
| 653 |
+
"num_input_tokens_seen": 19683060,
|
| 654 |
+
"step": 33000
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.077447574938891,
|
| 658 |
+
"grad_norm": 5.753154754638672,
|
| 659 |
+
"learning_rate": 3.204254041768515e-05,
|
| 660 |
+
"loss": 2.6977,
|
| 661 |
+
"num_input_tokens_seen": 19979196,
|
| 662 |
+
"step": 33500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 1.0935288820275313,
|
| 666 |
+
"grad_norm": 8.155505180358887,
|
| 667 |
+
"learning_rate": 3.1774518632874485e-05,
|
| 668 |
+
"loss": 2.7048,
|
| 669 |
+
"num_input_tokens_seen": 20278524,
|
| 670 |
+
"step": 34000
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 1.1096101891161714,
|
| 674 |
+
"grad_norm": 7.031659126281738,
|
| 675 |
+
"learning_rate": 3.150649684806381e-05,
|
| 676 |
+
"loss": 2.7237,
|
| 677 |
+
"num_input_tokens_seen": 20577572,
|
| 678 |
+
"step": 34500
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 1.1256914962048115,
|
| 682 |
+
"grad_norm": 7.90298318862915,
|
| 683 |
+
"learning_rate": 3.123847506325314e-05,
|
| 684 |
+
"loss": 2.7248,
|
| 685 |
+
"num_input_tokens_seen": 20876844,
|
| 686 |
+
"step": 35000
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 1.1256914962048115,
|
| 690 |
+
"eval_loss": 2.6041972637176514,
|
| 691 |
+
"eval_runtime": 2.7564,
|
| 692 |
+
"eval_samples_per_second": 906.969,
|
| 693 |
+
"eval_steps_per_second": 113.552,
|
| 694 |
+
"num_input_tokens_seen": 20876844,
|
| 695 |
+
"step": 35000
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.1417728032934518,
|
| 699 |
+
"grad_norm": 6.368433475494385,
|
| 700 |
+
"learning_rate": 3.0970453278442473e-05,
|
| 701 |
+
"loss": 2.7246,
|
| 702 |
+
"num_input_tokens_seen": 21179820,
|
| 703 |
+
"step": 35500
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.1578541103820919,
|
| 707 |
+
"grad_norm": 7.143220901489258,
|
| 708 |
+
"learning_rate": 3.0702431493631805e-05,
|
| 709 |
+
"loss": 2.7211,
|
| 710 |
+
"num_input_tokens_seen": 21476172,
|
| 711 |
+
"step": 36000
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 1.173935417470732,
|
| 715 |
+
"grad_norm": 7.216341972351074,
|
| 716 |
+
"learning_rate": 3.0434409708821134e-05,
|
| 717 |
+
"loss": 2.7088,
|
| 718 |
+
"num_input_tokens_seen": 21774292,
|
| 719 |
+
"step": 36500
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 1.1900167245593722,
|
| 723 |
+
"grad_norm": 6.958596706390381,
|
| 724 |
+
"learning_rate": 3.0166387924010465e-05,
|
| 725 |
+
"loss": 2.7166,
|
| 726 |
+
"num_input_tokens_seen": 22070908,
|
| 727 |
+
"step": 37000
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 1.2060980316480123,
|
| 731 |
+
"grad_norm": 7.161530494689941,
|
| 732 |
+
"learning_rate": 2.9898366139199797e-05,
|
| 733 |
+
"loss": 2.6764,
|
| 734 |
+
"num_input_tokens_seen": 22372340,
|
| 735 |
+
"step": 37500
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 1.2060980316480123,
|
| 739 |
+
"eval_loss": 2.5923423767089844,
|
| 740 |
+
"eval_runtime": 2.6849,
|
| 741 |
+
"eval_samples_per_second": 931.145,
|
| 742 |
+
"eval_steps_per_second": 116.579,
|
| 743 |
+
"num_input_tokens_seen": 22372340,
|
| 744 |
+
"step": 37500
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.2221793387366526,
|
| 748 |
+
"grad_norm": 7.448612213134766,
|
| 749 |
+
"learning_rate": 2.963034435438913e-05,
|
| 750 |
+
"loss": 2.7098,
|
| 751 |
+
"num_input_tokens_seen": 22672932,
|
| 752 |
+
"step": 38000
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.2382606458252927,
|
| 756 |
+
"grad_norm": 8.339189529418945,
|
| 757 |
+
"learning_rate": 2.9362322569578454e-05,
|
| 758 |
+
"loss": 2.702,
|
| 759 |
+
"num_input_tokens_seen": 22971844,
|
| 760 |
+
"step": 38500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 1.2543419529139328,
|
| 764 |
+
"grad_norm": 6.795124053955078,
|
| 765 |
+
"learning_rate": 2.9094300784767786e-05,
|
| 766 |
+
"loss": 2.7007,
|
| 767 |
+
"num_input_tokens_seen": 23266964,
|
| 768 |
+
"step": 39000
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 1.2704232600025729,
|
| 772 |
+
"grad_norm": 6.3036298751831055,
|
| 773 |
+
"learning_rate": 2.8826278999957118e-05,
|
| 774 |
+
"loss": 2.71,
|
| 775 |
+
"num_input_tokens_seen": 23564068,
|
| 776 |
+
"step": 39500
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 1.2865045670912132,
|
| 780 |
+
"grad_norm": 8.75069808959961,
|
| 781 |
+
"learning_rate": 2.855825721514645e-05,
|
| 782 |
+
"loss": 2.6854,
|
| 783 |
+
"num_input_tokens_seen": 23866100,
|
| 784 |
+
"step": 40000
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 1.2865045670912132,
|
| 788 |
+
"eval_loss": 2.5792863368988037,
|
| 789 |
+
"eval_runtime": 2.5776,
|
| 790 |
+
"eval_samples_per_second": 969.878,
|
| 791 |
+
"eval_steps_per_second": 121.429,
|
| 792 |
+
"num_input_tokens_seen": 23866100,
|
| 793 |
+
"step": 40000
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.3025858741798533,
|
| 797 |
+
"grad_norm": 6.966170310974121,
|
| 798 |
+
"learning_rate": 2.8290235430335778e-05,
|
| 799 |
+
"loss": 2.697,
|
| 800 |
+
"num_input_tokens_seen": 24162356,
|
| 801 |
+
"step": 40500
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 1.3186671812684936,
|
| 805 |
+
"grad_norm": 7.854964733123779,
|
| 806 |
+
"learning_rate": 2.802221364552511e-05,
|
| 807 |
+
"loss": 2.6954,
|
| 808 |
+
"num_input_tokens_seen": 24458980,
|
| 809 |
+
"step": 41000
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 1.3347484883571337,
|
| 813 |
+
"grad_norm": 7.1461944580078125,
|
| 814 |
+
"learning_rate": 2.775419186071444e-05,
|
| 815 |
+
"loss": 2.6839,
|
| 816 |
+
"num_input_tokens_seen": 24757828,
|
| 817 |
+
"step": 41500
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 1.3508297954457738,
|
| 821 |
+
"grad_norm": 8.25295639038086,
|
| 822 |
+
"learning_rate": 2.7486170075903773e-05,
|
| 823 |
+
"loss": 2.7035,
|
| 824 |
+
"num_input_tokens_seen": 25052236,
|
| 825 |
+
"step": 42000
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 1.366911102534414,
|
| 829 |
+
"grad_norm": 6.336223602294922,
|
| 830 |
+
"learning_rate": 2.7218148291093105e-05,
|
| 831 |
+
"loss": 2.683,
|
| 832 |
+
"num_input_tokens_seen": 25348084,
|
| 833 |
+
"step": 42500
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 1.366911102534414,
|
| 837 |
+
"eval_loss": 2.5722219944000244,
|
| 838 |
+
"eval_runtime": 2.7384,
|
| 839 |
+
"eval_samples_per_second": 912.958,
|
| 840 |
+
"eval_steps_per_second": 114.302,
|
| 841 |
+
"num_input_tokens_seen": 25348084,
|
| 842 |
+
"step": 42500
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.3829924096230541,
|
| 846 |
+
"grad_norm": 9.477555274963379,
|
| 847 |
+
"learning_rate": 2.695012650628243e-05,
|
| 848 |
+
"loss": 2.6877,
|
| 849 |
+
"num_input_tokens_seen": 25642372,
|
| 850 |
+
"step": 43000
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.3990737167116944,
|
| 854 |
+
"grad_norm": 8.233431816101074,
|
| 855 |
+
"learning_rate": 2.6682104721471762e-05,
|
| 856 |
+
"loss": 2.6927,
|
| 857 |
+
"num_input_tokens_seen": 25939652,
|
| 858 |
+
"step": 43500
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 1.4151550238003345,
|
| 862 |
+
"grad_norm": 5.860446929931641,
|
| 863 |
+
"learning_rate": 2.6414082936661094e-05,
|
| 864 |
+
"loss": 2.6819,
|
| 865 |
+
"num_input_tokens_seen": 26248940,
|
| 866 |
+
"step": 44000
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 1.4312363308889746,
|
| 870 |
+
"grad_norm": 6.748124599456787,
|
| 871 |
+
"learning_rate": 2.6146061151850426e-05,
|
| 872 |
+
"loss": 2.6893,
|
| 873 |
+
"num_input_tokens_seen": 26552860,
|
| 874 |
+
"step": 44500
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 1.4473176379776147,
|
| 878 |
+
"grad_norm": 6.038182258605957,
|
| 879 |
+
"learning_rate": 2.587803936703975e-05,
|
| 880 |
+
"loss": 2.6871,
|
| 881 |
+
"num_input_tokens_seen": 26854100,
|
| 882 |
+
"step": 45000
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 1.4473176379776147,
|
| 886 |
+
"eval_loss": 2.5538456439971924,
|
| 887 |
+
"eval_runtime": 2.6078,
|
| 888 |
+
"eval_samples_per_second": 958.67,
|
| 889 |
+
"eval_steps_per_second": 120.026,
|
| 890 |
+
"num_input_tokens_seen": 26854100,
|
| 891 |
+
"step": 45000
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.463398945066255,
|
| 895 |
+
"grad_norm": 7.815784454345703,
|
| 896 |
+
"learning_rate": 2.5610017582229086e-05,
|
| 897 |
+
"loss": 2.6709,
|
| 898 |
+
"num_input_tokens_seen": 27148148,
|
| 899 |
+
"step": 45500
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.479480252154895,
|
| 903 |
+
"grad_norm": 7.8851094245910645,
|
| 904 |
+
"learning_rate": 2.5341995797418418e-05,
|
| 905 |
+
"loss": 2.6698,
|
| 906 |
+
"num_input_tokens_seen": 27445020,
|
| 907 |
+
"step": 46000
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 1.4955615592435354,
|
| 911 |
+
"grad_norm": 7.389246940612793,
|
| 912 |
+
"learning_rate": 2.507397401260775e-05,
|
| 913 |
+
"loss": 2.6787,
|
| 914 |
+
"num_input_tokens_seen": 27742908,
|
| 915 |
+
"step": 46500
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 1.5116428663321755,
|
| 919 |
+
"grad_norm": 7.621913909912109,
|
| 920 |
+
"learning_rate": 2.4805952227797078e-05,
|
| 921 |
+
"loss": 2.6713,
|
| 922 |
+
"num_input_tokens_seen": 28037284,
|
| 923 |
+
"step": 47000
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 1.5277241734208156,
|
| 927 |
+
"grad_norm": 7.889066219329834,
|
| 928 |
+
"learning_rate": 2.4537930442986407e-05,
|
| 929 |
+
"loss": 2.6551,
|
| 930 |
+
"num_input_tokens_seen": 28332612,
|
| 931 |
+
"step": 47500
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 1.5277241734208156,
|
| 935 |
+
"eval_loss": 2.5442593097686768,
|
| 936 |
+
"eval_runtime": 2.6341,
|
| 937 |
+
"eval_samples_per_second": 949.086,
|
| 938 |
+
"eval_steps_per_second": 118.826,
|
| 939 |
+
"num_input_tokens_seen": 28332612,
|
| 940 |
+
"step": 47500
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.5438054805094557,
|
| 944 |
+
"grad_norm": 7.912906646728516,
|
| 945 |
+
"learning_rate": 2.426990865817574e-05,
|
| 946 |
+
"loss": 2.6881,
|
| 947 |
+
"num_input_tokens_seen": 28630948,
|
| 948 |
+
"step": 48000
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.559886787598096,
|
| 952 |
+
"grad_norm": 6.370878219604492,
|
| 953 |
+
"learning_rate": 2.4001886873365067e-05,
|
| 954 |
+
"loss": 2.6424,
|
| 955 |
+
"num_input_tokens_seen": 28928732,
|
| 956 |
+
"step": 48500
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 1.5759680946867363,
|
| 960 |
+
"grad_norm": 7.0892653465271,
|
| 961 |
+
"learning_rate": 2.37338650885544e-05,
|
| 962 |
+
"loss": 2.6626,
|
| 963 |
+
"num_input_tokens_seen": 29224436,
|
| 964 |
+
"step": 49000
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 1.5920494017753763,
|
| 968 |
+
"grad_norm": 6.357864856719971,
|
| 969 |
+
"learning_rate": 2.3465843303743727e-05,
|
| 970 |
+
"loss": 2.6546,
|
| 971 |
+
"num_input_tokens_seen": 29520148,
|
| 972 |
+
"step": 49500
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 1.6081307088640164,
|
| 976 |
+
"grad_norm": 8.4866943359375,
|
| 977 |
+
"learning_rate": 2.319782151893306e-05,
|
| 978 |
+
"loss": 2.661,
|
| 979 |
+
"num_input_tokens_seen": 29822156,
|
| 980 |
+
"step": 50000
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 1.6081307088640164,
|
| 984 |
+
"eval_loss": 2.527804374694824,
|
| 985 |
+
"eval_runtime": 2.6771,
|
| 986 |
+
"eval_samples_per_second": 933.855,
|
| 987 |
+
"eval_steps_per_second": 116.919,
|
| 988 |
+
"num_input_tokens_seen": 29822156,
|
| 989 |
+
"step": 50000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 1.6242120159526565,
|
| 993 |
+
"grad_norm": 6.843733787536621,
|
| 994 |
+
"learning_rate": 2.292979973412239e-05,
|
| 995 |
+
"loss": 2.6521,
|
| 996 |
+
"num_input_tokens_seen": 30122052,
|
| 997 |
+
"step": 50500
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.6402933230412968,
|
| 1001 |
+
"grad_norm": 6.88835334777832,
|
| 1002 |
+
"learning_rate": 2.2661777949311722e-05,
|
| 1003 |
+
"loss": 2.6614,
|
| 1004 |
+
"num_input_tokens_seen": 30422196,
|
| 1005 |
+
"step": 51000
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 1.656374630129937,
|
| 1009 |
+
"grad_norm": 5.855214595794678,
|
| 1010 |
+
"learning_rate": 2.239375616450105e-05,
|
| 1011 |
+
"loss": 2.6454,
|
| 1012 |
+
"num_input_tokens_seen": 30722660,
|
| 1013 |
+
"step": 51500
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 1.6724559372185772,
|
| 1017 |
+
"grad_norm": 6.58035135269165,
|
| 1018 |
+
"learning_rate": 2.2125734379690383e-05,
|
| 1019 |
+
"loss": 2.6524,
|
| 1020 |
+
"num_input_tokens_seen": 31018220,
|
| 1021 |
+
"step": 52000
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 1.6885372443072173,
|
| 1025 |
+
"grad_norm": 6.767495155334473,
|
| 1026 |
+
"learning_rate": 2.185771259487971e-05,
|
| 1027 |
+
"loss": 2.6497,
|
| 1028 |
+
"num_input_tokens_seen": 31319476,
|
| 1029 |
+
"step": 52500
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 1.6885372443072173,
|
| 1033 |
+
"eval_loss": 2.526638984680176,
|
| 1034 |
+
"eval_runtime": 2.5213,
|
| 1035 |
+
"eval_samples_per_second": 991.545,
|
| 1036 |
+
"eval_steps_per_second": 124.141,
|
| 1037 |
+
"num_input_tokens_seen": 31319476,
|
| 1038 |
+
"step": 52500
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 1.7046185513958574,
|
| 1042 |
+
"grad_norm": 7.022729873657227,
|
| 1043 |
+
"learning_rate": 2.1589690810069043e-05,
|
| 1044 |
+
"loss": 2.6437,
|
| 1045 |
+
"num_input_tokens_seen": 31621308,
|
| 1046 |
+
"step": 53000
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.7206998584844975,
|
| 1050 |
+
"grad_norm": 6.241069793701172,
|
| 1051 |
+
"learning_rate": 2.132166902525837e-05,
|
| 1052 |
+
"loss": 2.6447,
|
| 1053 |
+
"num_input_tokens_seen": 31917460,
|
| 1054 |
+
"step": 53500
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 1.7367811655731378,
|
| 1058 |
+
"grad_norm": 7.7204084396362305,
|
| 1059 |
+
"learning_rate": 2.1053647240447703e-05,
|
| 1060 |
+
"loss": 2.6448,
|
| 1061 |
+
"num_input_tokens_seen": 32217596,
|
| 1062 |
+
"step": 54000
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 1.752862472661778,
|
| 1066 |
+
"grad_norm": 6.703210830688477,
|
| 1067 |
+
"learning_rate": 2.0785625455637035e-05,
|
| 1068 |
+
"loss": 2.6366,
|
| 1069 |
+
"num_input_tokens_seen": 32513884,
|
| 1070 |
+
"step": 54500
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 1.7689437797504182,
|
| 1074 |
+
"grad_norm": 6.371466159820557,
|
| 1075 |
+
"learning_rate": 2.0517603670826367e-05,
|
| 1076 |
+
"loss": 2.6281,
|
| 1077 |
+
"num_input_tokens_seen": 32813220,
|
| 1078 |
+
"step": 55000
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 1.7689437797504182,
|
| 1082 |
+
"eval_loss": 2.5115973949432373,
|
| 1083 |
+
"eval_runtime": 2.5216,
|
| 1084 |
+
"eval_samples_per_second": 991.443,
|
| 1085 |
+
"eval_steps_per_second": 124.129,
|
| 1086 |
+
"num_input_tokens_seen": 32813220,
|
| 1087 |
+
"step": 55000
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 1.7850250868390583,
|
| 1091 |
+
"grad_norm": 7.277946949005127,
|
| 1092 |
+
"learning_rate": 2.0249581886015695e-05,
|
| 1093 |
+
"loss": 2.6536,
|
| 1094 |
+
"num_input_tokens_seen": 33110188,
|
| 1095 |
+
"step": 55500
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.8011063939276983,
|
| 1099 |
+
"grad_norm": 7.93104887008667,
|
| 1100 |
+
"learning_rate": 1.9981560101205027e-05,
|
| 1101 |
+
"loss": 2.5981,
|
| 1102 |
+
"num_input_tokens_seen": 33405980,
|
| 1103 |
+
"step": 56000
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 1.8171877010163386,
|
| 1107 |
+
"grad_norm": 7.782486438751221,
|
| 1108 |
+
"learning_rate": 1.971353831639436e-05,
|
| 1109 |
+
"loss": 2.635,
|
| 1110 |
+
"num_input_tokens_seen": 33700596,
|
| 1111 |
+
"step": 56500
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 1.8332690081049787,
|
| 1115 |
+
"grad_norm": 8.59358024597168,
|
| 1116 |
+
"learning_rate": 1.9445516531583687e-05,
|
| 1117 |
+
"loss": 2.6269,
|
| 1118 |
+
"num_input_tokens_seen": 33997724,
|
| 1119 |
+
"step": 57000
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 1.849350315193619,
|
| 1123 |
+
"grad_norm": 6.669950485229492,
|
| 1124 |
+
"learning_rate": 1.917749474677302e-05,
|
| 1125 |
+
"loss": 2.6067,
|
| 1126 |
+
"num_input_tokens_seen": 34298052,
|
| 1127 |
+
"step": 57500
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 1.849350315193619,
|
| 1131 |
+
"eval_loss": 2.5047078132629395,
|
| 1132 |
+
"eval_runtime": 2.5169,
|
| 1133 |
+
"eval_samples_per_second": 993.288,
|
| 1134 |
+
"eval_steps_per_second": 124.36,
|
| 1135 |
+
"num_input_tokens_seen": 34298052,
|
| 1136 |
+
"step": 57500
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 1.8654316222822591,
|
| 1140 |
+
"grad_norm": 6.265903949737549,
|
| 1141 |
+
"learning_rate": 1.8909472961962348e-05,
|
| 1142 |
+
"loss": 2.5966,
|
| 1143 |
+
"num_input_tokens_seen": 34593980,
|
| 1144 |
+
"step": 58000
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.8815129293708992,
|
| 1148 |
+
"grad_norm": 7.943974018096924,
|
| 1149 |
+
"learning_rate": 1.864145117715168e-05,
|
| 1150 |
+
"loss": 2.6303,
|
| 1151 |
+
"num_input_tokens_seen": 34894428,
|
| 1152 |
+
"step": 58500
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 1.8975942364595393,
|
| 1156 |
+
"grad_norm": 8.290629386901855,
|
| 1157 |
+
"learning_rate": 1.837342939234101e-05,
|
| 1158 |
+
"loss": 2.6303,
|
| 1159 |
+
"num_input_tokens_seen": 35193236,
|
| 1160 |
+
"step": 59000
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 1.9136755435481796,
|
| 1164 |
+
"grad_norm": 7.974947929382324,
|
| 1165 |
+
"learning_rate": 1.8105407607530343e-05,
|
| 1166 |
+
"loss": 2.6272,
|
| 1167 |
+
"num_input_tokens_seen": 35486796,
|
| 1168 |
+
"step": 59500
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 1.92975685063682,
|
| 1172 |
+
"grad_norm": 5.827637195587158,
|
| 1173 |
+
"learning_rate": 1.783738582271967e-05,
|
| 1174 |
+
"loss": 2.6112,
|
| 1175 |
+
"num_input_tokens_seen": 35783604,
|
| 1176 |
+
"step": 60000
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 1.92975685063682,
|
| 1180 |
+
"eval_loss": 2.4935405254364014,
|
| 1181 |
+
"eval_runtime": 2.5107,
|
| 1182 |
+
"eval_samples_per_second": 995.741,
|
| 1183 |
+
"eval_steps_per_second": 124.667,
|
| 1184 |
+
"num_input_tokens_seen": 35783604,
|
| 1185 |
+
"step": 60000
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 1.94583815772546,
|
| 1189 |
+
"grad_norm": 6.535378456115723,
|
| 1190 |
+
"learning_rate": 1.7569364037909003e-05,
|
| 1191 |
+
"loss": 2.6135,
|
| 1192 |
+
"num_input_tokens_seen": 36086620,
|
| 1193 |
+
"step": 60500
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.9619194648141,
|
| 1197 |
+
"grad_norm": 6.398725986480713,
|
| 1198 |
+
"learning_rate": 1.7301342253098332e-05,
|
| 1199 |
+
"loss": 2.602,
|
| 1200 |
+
"num_input_tokens_seen": 36386140,
|
| 1201 |
+
"step": 61000
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 1.9780007719027402,
|
| 1205 |
+
"grad_norm": 6.332113265991211,
|
| 1206 |
+
"learning_rate": 1.7033320468287664e-05,
|
| 1207 |
+
"loss": 2.6258,
|
| 1208 |
+
"num_input_tokens_seen": 36684308,
|
| 1209 |
+
"step": 61500
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 1.9940820789913805,
|
| 1213 |
+
"grad_norm": 7.8002753257751465,
|
| 1214 |
+
"learning_rate": 1.6765298683476992e-05,
|
| 1215 |
+
"loss": 2.6226,
|
| 1216 |
+
"num_input_tokens_seen": 36984724,
|
| 1217 |
+
"step": 62000
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 2.0101633860800208,
|
| 1221 |
+
"grad_norm": 6.9457011222839355,
|
| 1222 |
+
"learning_rate": 1.6497276898666324e-05,
|
| 1223 |
+
"loss": 2.5207,
|
| 1224 |
+
"num_input_tokens_seen": 37281092,
|
| 1225 |
+
"step": 62500
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 2.0101633860800208,
|
| 1229 |
+
"eval_loss": 2.4945950508117676,
|
| 1230 |
+
"eval_runtime": 2.5094,
|
| 1231 |
+
"eval_samples_per_second": 996.26,
|
| 1232 |
+
"eval_steps_per_second": 124.732,
|
| 1233 |
+
"num_input_tokens_seen": 37281092,
|
| 1234 |
+
"step": 62500
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 2.026244693168661,
|
| 1238 |
+
"grad_norm": 7.541498184204102,
|
| 1239 |
+
"learning_rate": 1.6229255113855656e-05,
|
| 1240 |
+
"loss": 2.4728,
|
| 1241 |
+
"num_input_tokens_seen": 37582300,
|
| 1242 |
+
"step": 63000
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 2.042326000257301,
|
| 1246 |
+
"grad_norm": 6.7798027992248535,
|
| 1247 |
+
"learning_rate": 1.5961233329044987e-05,
|
| 1248 |
+
"loss": 2.4539,
|
| 1249 |
+
"num_input_tokens_seen": 37880828,
|
| 1250 |
+
"step": 63500
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 2.058407307345941,
|
| 1254 |
+
"grad_norm": 7.033351898193359,
|
| 1255 |
+
"learning_rate": 1.5693211544234316e-05,
|
| 1256 |
+
"loss": 2.4467,
|
| 1257 |
+
"num_input_tokens_seen": 38181276,
|
| 1258 |
+
"step": 64000
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"epoch": 2.074488614434581,
|
| 1262 |
+
"grad_norm": 6.487890720367432,
|
| 1263 |
+
"learning_rate": 1.5425189759423648e-05,
|
| 1264 |
+
"loss": 2.4764,
|
| 1265 |
+
"num_input_tokens_seen": 38473348,
|
| 1266 |
+
"step": 64500
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 2.0905699215232216,
|
| 1270 |
+
"grad_norm": 6.955127716064453,
|
| 1271 |
+
"learning_rate": 1.5157167974612976e-05,
|
| 1272 |
+
"loss": 2.4799,
|
| 1273 |
+
"num_input_tokens_seen": 38768588,
|
| 1274 |
+
"step": 65000
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 2.0905699215232216,
|
| 1278 |
+
"eval_loss": 2.491555690765381,
|
| 1279 |
+
"eval_runtime": 2.5076,
|
| 1280 |
+
"eval_samples_per_second": 996.967,
|
| 1281 |
+
"eval_steps_per_second": 124.82,
|
| 1282 |
+
"num_input_tokens_seen": 38768588,
|
| 1283 |
+
"step": 65000
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 2.1066512286118617,
|
| 1287 |
+
"grad_norm": 6.78762674331665,
|
| 1288 |
+
"learning_rate": 1.4889146189802308e-05,
|
| 1289 |
+
"loss": 2.4726,
|
| 1290 |
+
"num_input_tokens_seen": 39067460,
|
| 1291 |
+
"step": 65500
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 2.122732535700502,
|
| 1295 |
+
"grad_norm": 7.199331283569336,
|
| 1296 |
+
"learning_rate": 1.4621124404991638e-05,
|
| 1297 |
+
"loss": 2.4562,
|
| 1298 |
+
"num_input_tokens_seen": 39360244,
|
| 1299 |
+
"step": 66000
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 2.138813842789142,
|
| 1303 |
+
"grad_norm": 7.353775501251221,
|
| 1304 |
+
"learning_rate": 1.435310262018097e-05,
|
| 1305 |
+
"loss": 2.4629,
|
| 1306 |
+
"num_input_tokens_seen": 39660020,
|
| 1307 |
+
"step": 66500
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 2.154895149877782,
|
| 1311 |
+
"grad_norm": 6.827337265014648,
|
| 1312 |
+
"learning_rate": 1.4085080835370298e-05,
|
| 1313 |
+
"loss": 2.4817,
|
| 1314 |
+
"num_input_tokens_seen": 39960476,
|
| 1315 |
+
"step": 67000
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 2.170976456966422,
|
| 1319 |
+
"grad_norm": 6.532020092010498,
|
| 1320 |
+
"learning_rate": 1.381705905055963e-05,
|
| 1321 |
+
"loss": 2.4727,
|
| 1322 |
+
"num_input_tokens_seen": 40252972,
|
| 1323 |
+
"step": 67500
|
| 1324 |
+
},
|
| 1325 |
+
{
|
| 1326 |
+
"epoch": 2.170976456966422,
|
| 1327 |
+
"eval_loss": 2.4865615367889404,
|
| 1328 |
+
"eval_runtime": 2.5372,
|
| 1329 |
+
"eval_samples_per_second": 985.335,
|
| 1330 |
+
"eval_steps_per_second": 123.364,
|
| 1331 |
+
"num_input_tokens_seen": 40252972,
|
| 1332 |
+
"step": 67500
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 2.1870577640550626,
|
| 1336 |
+
"grad_norm": 6.601158142089844,
|
| 1337 |
+
"learning_rate": 1.354903726574896e-05,
|
| 1338 |
+
"loss": 2.4666,
|
| 1339 |
+
"num_input_tokens_seen": 40553732,
|
| 1340 |
+
"step": 68000
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 2.2031390711437027,
|
| 1344 |
+
"grad_norm": 7.200645446777344,
|
| 1345 |
+
"learning_rate": 1.3281015480938292e-05,
|
| 1346 |
+
"loss": 2.4657,
|
| 1347 |
+
"num_input_tokens_seen": 40851196,
|
| 1348 |
+
"step": 68500
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 2.2192203782323427,
|
| 1352 |
+
"grad_norm": 8.067240715026855,
|
| 1353 |
+
"learning_rate": 1.301299369612762e-05,
|
| 1354 |
+
"loss": 2.4801,
|
| 1355 |
+
"num_input_tokens_seen": 41149276,
|
| 1356 |
+
"step": 69000
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 2.235301685320983,
|
| 1360 |
+
"grad_norm": 7.724194526672363,
|
| 1361 |
+
"learning_rate": 1.2744971911316952e-05,
|
| 1362 |
+
"loss": 2.4766,
|
| 1363 |
+
"num_input_tokens_seen": 41448540,
|
| 1364 |
+
"step": 69500
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 2.251382992409623,
|
| 1368 |
+
"grad_norm": 6.999200344085693,
|
| 1369 |
+
"learning_rate": 1.2476950126506282e-05,
|
| 1370 |
+
"loss": 2.4719,
|
| 1371 |
+
"num_input_tokens_seen": 41746300,
|
| 1372 |
+
"step": 70000
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"epoch": 2.251382992409623,
|
| 1376 |
+
"eval_loss": 2.476020097732544,
|
| 1377 |
+
"eval_runtime": 2.5166,
|
| 1378 |
+
"eval_samples_per_second": 993.407,
|
| 1379 |
+
"eval_steps_per_second": 124.375,
|
| 1380 |
+
"num_input_tokens_seen": 41746300,
|
| 1381 |
+
"step": 70000
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 2.267464299498263,
|
| 1385 |
+
"grad_norm": 6.666884899139404,
|
| 1386 |
+
"learning_rate": 1.2208928341695614e-05,
|
| 1387 |
+
"loss": 2.4771,
|
| 1388 |
+
"num_input_tokens_seen": 42042532,
|
| 1389 |
+
"step": 70500
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 2.2835456065869035,
|
| 1393 |
+
"grad_norm": 8.354509353637695,
|
| 1394 |
+
"learning_rate": 1.1940906556884944e-05,
|
| 1395 |
+
"loss": 2.4679,
|
| 1396 |
+
"num_input_tokens_seen": 42341628,
|
| 1397 |
+
"step": 71000
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 2.2996269136755436,
|
| 1401 |
+
"grad_norm": 8.39284610748291,
|
| 1402 |
+
"learning_rate": 1.1672884772074275e-05,
|
| 1403 |
+
"loss": 2.4597,
|
| 1404 |
+
"num_input_tokens_seen": 42642948,
|
| 1405 |
+
"step": 71500
|
| 1406 |
+
},
|
| 1407 |
+
{
|
| 1408 |
+
"epoch": 2.3157082207641837,
|
| 1409 |
+
"grad_norm": 7.233700275421143,
|
| 1410 |
+
"learning_rate": 1.1404862987263605e-05,
|
| 1411 |
+
"loss": 2.4592,
|
| 1412 |
+
"num_input_tokens_seen": 42941820,
|
| 1413 |
+
"step": 72000
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 2.331789527852824,
|
| 1417 |
+
"grad_norm": 7.843503475189209,
|
| 1418 |
+
"learning_rate": 1.1136841202452935e-05,
|
| 1419 |
+
"loss": 2.4738,
|
| 1420 |
+
"num_input_tokens_seen": 43241188,
|
| 1421 |
+
"step": 72500
|
| 1422 |
+
},
|
| 1423 |
+
{
|
| 1424 |
+
"epoch": 2.331789527852824,
|
| 1425 |
+
"eval_loss": 2.47127103805542,
|
| 1426 |
+
"eval_runtime": 2.5277,
|
| 1427 |
+
"eval_samples_per_second": 989.05,
|
| 1428 |
+
"eval_steps_per_second": 123.829,
|
| 1429 |
+
"num_input_tokens_seen": 43241188,
|
| 1430 |
+
"step": 72500
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 2.347870834941464,
|
| 1434 |
+
"grad_norm": 6.37482213973999,
|
| 1435 |
+
"learning_rate": 1.0868819417642267e-05,
|
| 1436 |
+
"loss": 2.4576,
|
| 1437 |
+
"num_input_tokens_seen": 43535900,
|
| 1438 |
+
"step": 73000
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 2.3639521420301044,
|
| 1442 |
+
"grad_norm": 6.642532825469971,
|
| 1443 |
+
"learning_rate": 1.0600797632831597e-05,
|
| 1444 |
+
"loss": 2.467,
|
| 1445 |
+
"num_input_tokens_seen": 43833516,
|
| 1446 |
+
"step": 73500
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 2.3800334491187445,
|
| 1450 |
+
"grad_norm": 6.606197357177734,
|
| 1451 |
+
"learning_rate": 1.0332775848020927e-05,
|
| 1452 |
+
"loss": 2.4752,
|
| 1453 |
+
"num_input_tokens_seen": 44134084,
|
| 1454 |
+
"step": 74000
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 2.3961147562073846,
|
| 1458 |
+
"grad_norm": 6.338978290557861,
|
| 1459 |
+
"learning_rate": 1.0064754063210257e-05,
|
| 1460 |
+
"loss": 2.4473,
|
| 1461 |
+
"num_input_tokens_seen": 44432540,
|
| 1462 |
+
"step": 74500
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"epoch": 2.4121960632960247,
|
| 1466 |
+
"grad_norm": 7.172792434692383,
|
| 1467 |
+
"learning_rate": 9.796732278399589e-06,
|
| 1468 |
+
"loss": 2.4629,
|
| 1469 |
+
"num_input_tokens_seen": 44730244,
|
| 1470 |
+
"step": 75000
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 2.4121960632960247,
|
| 1474 |
+
"eval_loss": 2.4629955291748047,
|
| 1475 |
+
"eval_runtime": 2.521,
|
| 1476 |
+
"eval_samples_per_second": 991.68,
|
| 1477 |
+
"eval_steps_per_second": 124.158,
|
| 1478 |
+
"num_input_tokens_seen": 44730244,
|
| 1479 |
+
"step": 75000
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 2.4282773703846647,
|
| 1483 |
+
"grad_norm": 6.65930700302124,
|
| 1484 |
+
"learning_rate": 9.528710493588919e-06,
|
| 1485 |
+
"loss": 2.4512,
|
| 1486 |
+
"num_input_tokens_seen": 45031884,
|
| 1487 |
+
"step": 75500
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 2.4443586774733053,
|
| 1491 |
+
"grad_norm": 7.209745407104492,
|
| 1492 |
+
"learning_rate": 9.260688708778249e-06,
|
| 1493 |
+
"loss": 2.4557,
|
| 1494 |
+
"num_input_tokens_seen": 45334924,
|
| 1495 |
+
"step": 76000
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 2.4604399845619453,
|
| 1499 |
+
"grad_norm": 6.847073078155518,
|
| 1500 |
+
"learning_rate": 8.992666923967581e-06,
|
| 1501 |
+
"loss": 2.4512,
|
| 1502 |
+
"num_input_tokens_seen": 45633532,
|
| 1503 |
+
"step": 76500
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"epoch": 2.4765212916505854,
|
| 1507 |
+
"grad_norm": 7.705162525177002,
|
| 1508 |
+
"learning_rate": 8.724645139156911e-06,
|
| 1509 |
+
"loss": 2.4568,
|
| 1510 |
+
"num_input_tokens_seen": 45933804,
|
| 1511 |
+
"step": 77000
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 2.4926025987392255,
|
| 1515 |
+
"grad_norm": 7.5681962966918945,
|
| 1516 |
+
"learning_rate": 8.456623354346243e-06,
|
| 1517 |
+
"loss": 2.4524,
|
| 1518 |
+
"num_input_tokens_seen": 46231060,
|
| 1519 |
+
"step": 77500
|
| 1520 |
+
},
|
| 1521 |
+
{
|
| 1522 |
+
"epoch": 2.4926025987392255,
|
| 1523 |
+
"eval_loss": 2.457481861114502,
|
| 1524 |
+
"eval_runtime": 2.5921,
|
| 1525 |
+
"eval_samples_per_second": 964.467,
|
| 1526 |
+
"eval_steps_per_second": 120.751,
|
| 1527 |
+
"num_input_tokens_seen": 46231060,
|
| 1528 |
+
"step": 77500
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 2.5086839058278656,
|
| 1532 |
+
"grad_norm": 6.8857269287109375,
|
| 1533 |
+
"learning_rate": 8.188601569535573e-06,
|
| 1534 |
+
"loss": 2.4622,
|
| 1535 |
+
"num_input_tokens_seen": 46525772,
|
| 1536 |
+
"step": 78000
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 2.524765212916506,
|
| 1540 |
+
"grad_norm": 6.347681522369385,
|
| 1541 |
+
"learning_rate": 7.920579784724903e-06,
|
| 1542 |
+
"loss": 2.4528,
|
| 1543 |
+
"num_input_tokens_seen": 46822532,
|
| 1544 |
+
"step": 78500
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 2.5408465200051458,
|
| 1548 |
+
"grad_norm": 6.935575008392334,
|
| 1549 |
+
"learning_rate": 7.652557999914233e-06,
|
| 1550 |
+
"loss": 2.4414,
|
| 1551 |
+
"num_input_tokens_seen": 47122964,
|
| 1552 |
+
"step": 79000
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 2.5569278270937863,
|
| 1556 |
+
"grad_norm": 6.603360652923584,
|
| 1557 |
+
"learning_rate": 7.384536215103564e-06,
|
| 1558 |
+
"loss": 2.4655,
|
| 1559 |
+
"num_input_tokens_seen": 47423300,
|
| 1560 |
+
"step": 79500
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 2.5730091341824264,
|
| 1564 |
+
"grad_norm": 7.182071208953857,
|
| 1565 |
+
"learning_rate": 7.116514430292895e-06,
|
| 1566 |
+
"loss": 2.435,
|
| 1567 |
+
"num_input_tokens_seen": 47718964,
|
| 1568 |
+
"step": 80000
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 2.5730091341824264,
|
| 1572 |
+
"eval_loss": 2.455320358276367,
|
| 1573 |
+
"eval_runtime": 2.5065,
|
| 1574 |
+
"eval_samples_per_second": 997.4,
|
| 1575 |
+
"eval_steps_per_second": 124.874,
|
| 1576 |
+
"num_input_tokens_seen": 47718964,
|
| 1577 |
+
"step": 80000
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 2.5890904412710665,
|
| 1581 |
+
"grad_norm": 7.3647260665893555,
|
| 1582 |
+
"learning_rate": 6.848492645482225e-06,
|
| 1583 |
+
"loss": 2.4356,
|
| 1584 |
+
"num_input_tokens_seen": 48015996,
|
| 1585 |
+
"step": 80500
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 2.6051717483597066,
|
| 1589 |
+
"grad_norm": 7.950341701507568,
|
| 1590 |
+
"learning_rate": 6.580470860671556e-06,
|
| 1591 |
+
"loss": 2.4453,
|
| 1592 |
+
"num_input_tokens_seen": 48316420,
|
| 1593 |
+
"step": 81000
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 2.6212530554483466,
|
| 1597 |
+
"grad_norm": 6.016787052154541,
|
| 1598 |
+
"learning_rate": 6.312449075860886e-06,
|
| 1599 |
+
"loss": 2.45,
|
| 1600 |
+
"num_input_tokens_seen": 48611452,
|
| 1601 |
+
"step": 81500
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 2.637334362536987,
|
| 1605 |
+
"grad_norm": 7.281980514526367,
|
| 1606 |
+
"learning_rate": 6.044427291050217e-06,
|
| 1607 |
+
"loss": 2.4687,
|
| 1608 |
+
"num_input_tokens_seen": 48913668,
|
| 1609 |
+
"step": 82000
|
| 1610 |
+
},
|
| 1611 |
+
{
|
| 1612 |
+
"epoch": 2.6534156696256272,
|
| 1613 |
+
"grad_norm": 6.644787311553955,
|
| 1614 |
+
"learning_rate": 5.776405506239547e-06,
|
| 1615 |
+
"loss": 2.4621,
|
| 1616 |
+
"num_input_tokens_seen": 49209724,
|
| 1617 |
+
"step": 82500
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 2.6534156696256272,
|
| 1621 |
+
"eval_loss": 2.4475488662719727,
|
| 1622 |
+
"eval_runtime": 2.517,
|
| 1623 |
+
"eval_samples_per_second": 993.239,
|
| 1624 |
+
"eval_steps_per_second": 124.354,
|
| 1625 |
+
"num_input_tokens_seen": 49209724,
|
| 1626 |
+
"step": 82500
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 2.6694969767142673,
|
| 1630 |
+
"grad_norm": 6.181220054626465,
|
| 1631 |
+
"learning_rate": 5.508383721428878e-06,
|
| 1632 |
+
"loss": 2.4343,
|
| 1633 |
+
"num_input_tokens_seen": 49505772,
|
| 1634 |
+
"step": 83000
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 2.6855782838029074,
|
| 1638 |
+
"grad_norm": 6.418393135070801,
|
| 1639 |
+
"learning_rate": 5.2403619366182085e-06,
|
| 1640 |
+
"loss": 2.4329,
|
| 1641 |
+
"num_input_tokens_seen": 49809956,
|
| 1642 |
+
"step": 83500
|
| 1643 |
+
},
|
| 1644 |
+
{
|
| 1645 |
+
"epoch": 2.7016595908915475,
|
| 1646 |
+
"grad_norm": 6.279716491699219,
|
| 1647 |
+
"learning_rate": 4.9723401518075395e-06,
|
| 1648 |
+
"loss": 2.4481,
|
| 1649 |
+
"num_input_tokens_seen": 50112060,
|
| 1650 |
+
"step": 84000
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 2.717740897980188,
|
| 1654 |
+
"grad_norm": 6.502873420715332,
|
| 1655 |
+
"learning_rate": 4.70431836699687e-06,
|
| 1656 |
+
"loss": 2.4464,
|
| 1657 |
+
"num_input_tokens_seen": 50414356,
|
| 1658 |
+
"step": 84500
|
| 1659 |
+
},
|
| 1660 |
+
{
|
| 1661 |
+
"epoch": 2.733822205068828,
|
| 1662 |
+
"grad_norm": 6.15990686416626,
|
| 1663 |
+
"learning_rate": 4.4362965821862e-06,
|
| 1664 |
+
"loss": 2.4492,
|
| 1665 |
+
"num_input_tokens_seen": 50712980,
|
| 1666 |
+
"step": 85000
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 2.733822205068828,
|
| 1670 |
+
"eval_loss": 2.4440150260925293,
|
| 1671 |
+
"eval_runtime": 2.529,
|
| 1672 |
+
"eval_samples_per_second": 988.544,
|
| 1673 |
+
"eval_steps_per_second": 123.766,
|
| 1674 |
+
"num_input_tokens_seen": 50712980,
|
| 1675 |
+
"step": 85000
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 2.749903512157468,
|
| 1679 |
+
"grad_norm": 6.876352310180664,
|
| 1680 |
+
"learning_rate": 4.168274797375531e-06,
|
| 1681 |
+
"loss": 2.4514,
|
| 1682 |
+
"num_input_tokens_seen": 51012460,
|
| 1683 |
+
"step": 85500
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 2.7659848192461083,
|
| 1687 |
+
"grad_norm": 7.305426597595215,
|
| 1688 |
+
"learning_rate": 3.900253012564861e-06,
|
| 1689 |
+
"loss": 2.4317,
|
| 1690 |
+
"num_input_tokens_seen": 51308524,
|
| 1691 |
+
"step": 86000
|
| 1692 |
+
},
|
| 1693 |
+
{
|
| 1694 |
+
"epoch": 2.7820661263347484,
|
| 1695 |
+
"grad_norm": 6.460892677307129,
|
| 1696 |
+
"learning_rate": 3.632231227754192e-06,
|
| 1697 |
+
"loss": 2.4559,
|
| 1698 |
+
"num_input_tokens_seen": 51610700,
|
| 1699 |
+
"step": 86500
|
| 1700 |
+
},
|
| 1701 |
+
{
|
| 1702 |
+
"epoch": 2.798147433423389,
|
| 1703 |
+
"grad_norm": 8.062651634216309,
|
| 1704 |
+
"learning_rate": 3.3642094429435228e-06,
|
| 1705 |
+
"loss": 2.4535,
|
| 1706 |
+
"num_input_tokens_seen": 51910236,
|
| 1707 |
+
"step": 87000
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 2.814228740512029,
|
| 1711 |
+
"grad_norm": 7.140311241149902,
|
| 1712 |
+
"learning_rate": 3.0961876581328533e-06,
|
| 1713 |
+
"loss": 2.4536,
|
| 1714 |
+
"num_input_tokens_seen": 52204380,
|
| 1715 |
+
"step": 87500
|
| 1716 |
+
},
|
| 1717 |
+
{
|
| 1718 |
+
"epoch": 2.814228740512029,
|
| 1719 |
+
"eval_loss": 2.4393906593322754,
|
| 1720 |
+
"eval_runtime": 2.5312,
|
| 1721 |
+
"eval_samples_per_second": 987.685,
|
| 1722 |
+
"eval_steps_per_second": 123.658,
|
| 1723 |
+
"num_input_tokens_seen": 52204380,
|
| 1724 |
+
"step": 87500
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 2.830310047600669,
|
| 1728 |
+
"grad_norm": 6.569787502288818,
|
| 1729 |
+
"learning_rate": 2.8281658733221834e-06,
|
| 1730 |
+
"loss": 2.4379,
|
| 1731 |
+
"num_input_tokens_seen": 52504668,
|
| 1732 |
+
"step": 88000
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 2.846391354689309,
|
| 1736 |
+
"grad_norm": 7.735711097717285,
|
| 1737 |
+
"learning_rate": 2.560144088511514e-06,
|
| 1738 |
+
"loss": 2.4239,
|
| 1739 |
+
"num_input_tokens_seen": 52798740,
|
| 1740 |
+
"step": 88500
|
| 1741 |
+
},
|
| 1742 |
+
{
|
| 1743 |
+
"epoch": 2.8624726617779492,
|
| 1744 |
+
"grad_norm": 7.504124641418457,
|
| 1745 |
+
"learning_rate": 2.292122303700845e-06,
|
| 1746 |
+
"loss": 2.4427,
|
| 1747 |
+
"num_input_tokens_seen": 53097716,
|
| 1748 |
+
"step": 89000
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"epoch": 2.8785539688665893,
|
| 1752 |
+
"grad_norm": 6.647756099700928,
|
| 1753 |
+
"learning_rate": 2.0241005188901755e-06,
|
| 1754 |
+
"loss": 2.4682,
|
| 1755 |
+
"num_input_tokens_seen": 53397564,
|
| 1756 |
+
"step": 89500
|
| 1757 |
+
},
|
| 1758 |
+
{
|
| 1759 |
+
"epoch": 2.8946352759552294,
|
| 1760 |
+
"grad_norm": 6.640815734863281,
|
| 1761 |
+
"learning_rate": 1.756078734079506e-06,
|
| 1762 |
+
"loss": 2.4148,
|
| 1763 |
+
"num_input_tokens_seen": 53695620,
|
| 1764 |
+
"step": 90000
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 2.8946352759552294,
|
| 1768 |
+
"eval_loss": 2.43597674369812,
|
| 1769 |
+
"eval_runtime": 2.5128,
|
| 1770 |
+
"eval_samples_per_second": 994.891,
|
| 1771 |
+
"eval_steps_per_second": 124.56,
|
| 1772 |
+
"num_input_tokens_seen": 53695620,
|
| 1773 |
+
"step": 90000
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 2.91071658304387,
|
| 1777 |
+
"grad_norm": 7.346447467803955,
|
| 1778 |
+
"learning_rate": 1.4880569492688366e-06,
|
| 1779 |
+
"loss": 2.4352,
|
| 1780 |
+
"num_input_tokens_seen": 53991180,
|
| 1781 |
+
"step": 90500
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 2.92679789013251,
|
| 1785 |
+
"grad_norm": 6.777767658233643,
|
| 1786 |
+
"learning_rate": 1.2200351644581672e-06,
|
| 1787 |
+
"loss": 2.4664,
|
| 1788 |
+
"num_input_tokens_seen": 54288348,
|
| 1789 |
+
"step": 91000
|
| 1790 |
+
},
|
| 1791 |
+
{
|
| 1792 |
+
"epoch": 2.94287919722115,
|
| 1793 |
+
"grad_norm": 6.908254623413086,
|
| 1794 |
+
"learning_rate": 9.520133796474978e-07,
|
| 1795 |
+
"loss": 2.4474,
|
| 1796 |
+
"num_input_tokens_seen": 54590740,
|
| 1797 |
+
"step": 91500
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"epoch": 2.95896050430979,
|
| 1801 |
+
"grad_norm": 7.04544734954834,
|
| 1802 |
+
"learning_rate": 6.839915948368284e-07,
|
| 1803 |
+
"loss": 2.4554,
|
| 1804 |
+
"num_input_tokens_seen": 54889220,
|
| 1805 |
+
"step": 92000
|
| 1806 |
+
},
|
| 1807 |
+
{
|
| 1808 |
+
"epoch": 2.9750418113984303,
|
| 1809 |
+
"grad_norm": 9.98161792755127,
|
| 1810 |
+
"learning_rate": 4.159698100261589e-07,
|
| 1811 |
+
"loss": 2.4243,
|
| 1812 |
+
"num_input_tokens_seen": 55190020,
|
| 1813 |
+
"step": 92500
|
| 1814 |
+
},
|
| 1815 |
+
{
|
| 1816 |
+
"epoch": 2.9750418113984303,
|
| 1817 |
+
"eval_loss": 2.435030460357666,
|
| 1818 |
+
"eval_runtime": 2.5128,
|
| 1819 |
+
"eval_samples_per_second": 994.908,
|
| 1820 |
+
"eval_steps_per_second": 124.563,
|
| 1821 |
+
"num_input_tokens_seen": 55190020,
|
| 1822 |
+
"step": 92500
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 2.991123118487071,
|
| 1826 |
+
"grad_norm": 6.586206912994385,
|
| 1827 |
+
"learning_rate": 1.479480252154895e-07,
|
| 1828 |
+
"loss": 2.44,
|
| 1829 |
+
"num_input_tokens_seen": 55490868,
|
| 1830 |
+
"step": 93000
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 3.0,
|
| 1834 |
+
"num_input_tokens_seen": 55653732,
|
| 1835 |
+
"step": 93276,
|
| 1836 |
+
"total_flos": 1.4738832163602432e+16,
|
| 1837 |
+
"train_loss": 1.7073542784139526,
|
| 1838 |
+
"train_runtime": 2504.3889,
|
| 1839 |
+
"train_samples_per_second": 297.955,
|
| 1840 |
+
"train_steps_per_second": 37.245,
|
| 1841 |
+
"train_tokens_per_second": 22216.164
|
| 1842 |
+
}
|
| 1843 |
+
],
|
| 1844 |
+
"logging_steps": 500,
|
| 1845 |
+
"max_steps": 93276,
|
| 1846 |
+
"num_input_tokens_seen": 55653732,
|
| 1847 |
+
"num_train_epochs": 3,
|
| 1848 |
+
"save_steps": 2500,
|
| 1849 |
+
"stateful_callbacks": {
|
| 1850 |
+
"TrainerControl": {
|
| 1851 |
+
"args": {
|
| 1852 |
+
"should_epoch_stop": false,
|
| 1853 |
+
"should_evaluate": false,
|
| 1854 |
+
"should_log": false,
|
| 1855 |
+
"should_save": true,
|
| 1856 |
+
"should_training_stop": true
|
| 1857 |
+
},
|
| 1858 |
+
"attributes": {}
|
| 1859 |
+
}
|
| 1860 |
+
},
|
| 1861 |
+
"total_flos": 1.4738832163602432e+16,
|
| 1862 |
+
"train_batch_size": 8,
|
| 1863 |
+
"trial_name": null,
|
| 1864 |
+
"trial_params": null
|
| 1865 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8b6b0e3aff2a6ecffae867eef3409d7416d0e3a25aa9423ea38026d8d618f479
|
| 3 |
+
size 5432
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|