Upload folder using huggingface_hub
Browse files- config.json +25 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +51 -0
- tokenizer.json +0 -0
- tokenizer_config.json +71 -0
- trainer_state.json +3634 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForMaskedLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"dtype": "float32",
|
| 8 |
+
"gradient_checkpointing": false,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 512,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"layer_norm_eps": 1e-12,
|
| 15 |
+
"max_position_embeddings": 1024,
|
| 16 |
+
"model_type": "bert",
|
| 17 |
+
"num_attention_heads": 8,
|
| 18 |
+
"num_hidden_layers": 4,
|
| 19 |
+
"pad_token_id": 0,
|
| 20 |
+
"position_embedding_type": "absolute",
|
| 21 |
+
"transformers_version": "4.57.6",
|
| 22 |
+
"type_vocab_size": 2,
|
| 23 |
+
"use_cache": true,
|
| 24 |
+
"vocab_size": 30522
|
| 25 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c48972ed56f41bc6850e69deb527e98193d23b540cf19aac207fb72e991efd02
|
| 3 |
+
size 133031496
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:acf84ed9905aecbeec8c15959433d21386c70135f7b8a92f08f69d84b70a2325
|
| 3 |
+
size 266109515
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cd2b6e658410dcd7def6a5026dcc483109a654719c408d809f1efb4f52967be0
|
| 3 |
+
size 14645
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cb99b5a4ab1433a51efe856b4ea1e15de233cd524729b809a7ab2bddd71c4e90
|
| 3 |
+
size 1465
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "[BOS]",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "[CLS]",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "[EOS]",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "[MASK]",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"pad_token": {
|
| 31 |
+
"content": "[PAD]",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"sep_token": {
|
| 38 |
+
"content": "[SEP]",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"unk_token": {
|
| 45 |
+
"content": "[UNK]",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": false,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,71 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[UNK]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "[CLS]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "[SEP]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "[PAD]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"4": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"30001": {
|
| 44 |
+
"content": "[EOS]",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"30002": {
|
| 52 |
+
"content": "[BOS]",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
}
|
| 59 |
+
},
|
| 60 |
+
"bos_token": "[BOS]",
|
| 61 |
+
"clean_up_tokenization_spaces": false,
|
| 62 |
+
"cls_token": "[CLS]",
|
| 63 |
+
"eos_token": "[EOS]",
|
| 64 |
+
"extra_special_tokens": {},
|
| 65 |
+
"mask_token": "[MASK]",
|
| 66 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 67 |
+
"pad_token": "[PAD]",
|
| 68 |
+
"sep_token": "[SEP]",
|
| 69 |
+
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 70 |
+
"unk_token": "[UNK]"
|
| 71 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,3634 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 12000,
|
| 3 |
+
"best_metric": 0.12119368463754654,
|
| 4 |
+
"best_model_checkpoint": "/data/alamparan/mattext_ckpt/results/2026-02-04/03-04-44/pretrain/checkpoints/robocrys_rep_test-pretrain/checkpoint-12000",
|
| 5 |
+
"epoch": 38.70967741935484,
|
| 6 |
+
"eval_steps": 50,
|
| 7 |
+
"global_step": 12000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.16129032258064516,
|
| 14 |
+
"grad_norm": 1.4281898736953735,
|
| 15 |
+
"learning_rate": 0.00019936774193548388,
|
| 16 |
+
"loss": 6.0329,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.16129032258064516,
|
| 21 |
+
"eval_loss": 4.073571681976318,
|
| 22 |
+
"eval_runtime": 86.5227,
|
| 23 |
+
"eval_samples_per_second": 219.642,
|
| 24 |
+
"eval_steps_per_second": 4.577,
|
| 25 |
+
"step": 50
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.3225806451612903,
|
| 29 |
+
"grad_norm": 1.1555994749069214,
|
| 30 |
+
"learning_rate": 0.0001987225806451613,
|
| 31 |
+
"loss": 3.8465,
|
| 32 |
+
"step": 100
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.3225806451612903,
|
| 36 |
+
"eval_loss": 3.6210274696350098,
|
| 37 |
+
"eval_runtime": 87.4418,
|
| 38 |
+
"eval_samples_per_second": 217.333,
|
| 39 |
+
"eval_steps_per_second": 4.529,
|
| 40 |
+
"step": 100
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.4838709677419355,
|
| 44 |
+
"grad_norm": 1.2039262056350708,
|
| 45 |
+
"learning_rate": 0.00019807741935483873,
|
| 46 |
+
"loss": 3.5783,
|
| 47 |
+
"step": 150
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.4838709677419355,
|
| 51 |
+
"eval_loss": 3.4356088638305664,
|
| 52 |
+
"eval_runtime": 86.395,
|
| 53 |
+
"eval_samples_per_second": 219.966,
|
| 54 |
+
"eval_steps_per_second": 4.584,
|
| 55 |
+
"step": 150
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.6451612903225806,
|
| 59 |
+
"grad_norm": 0.8887826204299927,
|
| 60 |
+
"learning_rate": 0.00019743225806451612,
|
| 61 |
+
"loss": 3.411,
|
| 62 |
+
"step": 200
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.6451612903225806,
|
| 66 |
+
"eval_loss": 3.294156074523926,
|
| 67 |
+
"eval_runtime": 89.1633,
|
| 68 |
+
"eval_samples_per_second": 213.137,
|
| 69 |
+
"eval_steps_per_second": 4.441,
|
| 70 |
+
"step": 200
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.8064516129032258,
|
| 74 |
+
"grad_norm": 0.9355249404907227,
|
| 75 |
+
"learning_rate": 0.00019678709677419356,
|
| 76 |
+
"loss": 3.3012,
|
| 77 |
+
"step": 250
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.8064516129032258,
|
| 81 |
+
"eval_loss": 3.1778066158294678,
|
| 82 |
+
"eval_runtime": 87.9303,
|
| 83 |
+
"eval_samples_per_second": 216.126,
|
| 84 |
+
"eval_steps_per_second": 4.504,
|
| 85 |
+
"step": 250
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.967741935483871,
|
| 89 |
+
"grad_norm": 0.9125154614448547,
|
| 90 |
+
"learning_rate": 0.000196141935483871,
|
| 91 |
+
"loss": 3.1839,
|
| 92 |
+
"step": 300
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.967741935483871,
|
| 96 |
+
"eval_loss": 3.0771267414093018,
|
| 97 |
+
"eval_runtime": 88.1645,
|
| 98 |
+
"eval_samples_per_second": 215.552,
|
| 99 |
+
"eval_steps_per_second": 4.492,
|
| 100 |
+
"step": 300
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 1.129032258064516,
|
| 104 |
+
"grad_norm": 1.038801670074463,
|
| 105 |
+
"learning_rate": 0.00019549677419354838,
|
| 106 |
+
"loss": 3.0949,
|
| 107 |
+
"step": 350
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.129032258064516,
|
| 111 |
+
"eval_loss": 2.972693920135498,
|
| 112 |
+
"eval_runtime": 87.3627,
|
| 113 |
+
"eval_samples_per_second": 217.53,
|
| 114 |
+
"eval_steps_per_second": 4.533,
|
| 115 |
+
"step": 350
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 1.2903225806451613,
|
| 119 |
+
"grad_norm": 0.9201057553291321,
|
| 120 |
+
"learning_rate": 0.00019485161290322582,
|
| 121 |
+
"loss": 3.0067,
|
| 122 |
+
"step": 400
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 1.2903225806451613,
|
| 126 |
+
"eval_loss": 2.8708949089050293,
|
| 127 |
+
"eval_runtime": 86.2708,
|
| 128 |
+
"eval_samples_per_second": 220.283,
|
| 129 |
+
"eval_steps_per_second": 4.59,
|
| 130 |
+
"step": 400
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.4516129032258065,
|
| 134 |
+
"grad_norm": 1.1271328926086426,
|
| 135 |
+
"learning_rate": 0.00019420645161290323,
|
| 136 |
+
"loss": 2.8894,
|
| 137 |
+
"step": 450
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.4516129032258065,
|
| 141 |
+
"eval_loss": 2.741400957107544,
|
| 142 |
+
"eval_runtime": 88.3865,
|
| 143 |
+
"eval_samples_per_second": 215.01,
|
| 144 |
+
"eval_steps_per_second": 4.48,
|
| 145 |
+
"step": 450
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 1.6129032258064515,
|
| 149 |
+
"grad_norm": 1.6507548093795776,
|
| 150 |
+
"learning_rate": 0.00019356129032258065,
|
| 151 |
+
"loss": 2.7396,
|
| 152 |
+
"step": 500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 1.6129032258064515,
|
| 156 |
+
"eval_loss": 2.5605974197387695,
|
| 157 |
+
"eval_runtime": 88.4696,
|
| 158 |
+
"eval_samples_per_second": 214.808,
|
| 159 |
+
"eval_steps_per_second": 4.476,
|
| 160 |
+
"step": 500
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 1.7741935483870968,
|
| 164 |
+
"grad_norm": 1.5166431665420532,
|
| 165 |
+
"learning_rate": 0.00019291612903225806,
|
| 166 |
+
"loss": 2.5539,
|
| 167 |
+
"step": 550
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 1.7741935483870968,
|
| 171 |
+
"eval_loss": 2.2101876735687256,
|
| 172 |
+
"eval_runtime": 88.7721,
|
| 173 |
+
"eval_samples_per_second": 214.076,
|
| 174 |
+
"eval_steps_per_second": 4.461,
|
| 175 |
+
"step": 550
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 1.935483870967742,
|
| 179 |
+
"grad_norm": 1.8889034986495972,
|
| 180 |
+
"learning_rate": 0.0001922709677419355,
|
| 181 |
+
"loss": 2.043,
|
| 182 |
+
"step": 600
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.935483870967742,
|
| 186 |
+
"eval_loss": 1.4450961351394653,
|
| 187 |
+
"eval_runtime": 87.3615,
|
| 188 |
+
"eval_samples_per_second": 217.533,
|
| 189 |
+
"eval_steps_per_second": 4.533,
|
| 190 |
+
"step": 600
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 2.096774193548387,
|
| 194 |
+
"grad_norm": 1.4880077838897705,
|
| 195 |
+
"learning_rate": 0.0001916258064516129,
|
| 196 |
+
"loss": 1.4473,
|
| 197 |
+
"step": 650
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 2.096774193548387,
|
| 201 |
+
"eval_loss": 1.047796607017517,
|
| 202 |
+
"eval_runtime": 88.4068,
|
| 203 |
+
"eval_samples_per_second": 214.961,
|
| 204 |
+
"eval_steps_per_second": 4.479,
|
| 205 |
+
"step": 650
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 2.258064516129032,
|
| 209 |
+
"grad_norm": 1.2545322179794312,
|
| 210 |
+
"learning_rate": 0.00019098064516129032,
|
| 211 |
+
"loss": 1.1162,
|
| 212 |
+
"step": 700
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 2.258064516129032,
|
| 216 |
+
"eval_loss": 0.8680551052093506,
|
| 217 |
+
"eval_runtime": 89.0396,
|
| 218 |
+
"eval_samples_per_second": 213.433,
|
| 219 |
+
"eval_steps_per_second": 4.447,
|
| 220 |
+
"step": 700
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 2.4193548387096775,
|
| 224 |
+
"grad_norm": 1.1532652378082275,
|
| 225 |
+
"learning_rate": 0.00019033548387096776,
|
| 226 |
+
"loss": 0.9731,
|
| 227 |
+
"step": 750
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 2.4193548387096775,
|
| 231 |
+
"eval_loss": 0.785129964351654,
|
| 232 |
+
"eval_runtime": 90.5213,
|
| 233 |
+
"eval_samples_per_second": 209.939,
|
| 234 |
+
"eval_steps_per_second": 4.375,
|
| 235 |
+
"step": 750
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 2.5806451612903225,
|
| 239 |
+
"grad_norm": 0.8688052892684937,
|
| 240 |
+
"learning_rate": 0.00018969032258064517,
|
| 241 |
+
"loss": 0.8907,
|
| 242 |
+
"step": 800
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 2.5806451612903225,
|
| 246 |
+
"eval_loss": 0.7259724140167236,
|
| 247 |
+
"eval_runtime": 88.9062,
|
| 248 |
+
"eval_samples_per_second": 213.753,
|
| 249 |
+
"eval_steps_per_second": 4.454,
|
| 250 |
+
"step": 800
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 2.741935483870968,
|
| 254 |
+
"grad_norm": 1.1574801206588745,
|
| 255 |
+
"learning_rate": 0.00018904516129032259,
|
| 256 |
+
"loss": 0.8301,
|
| 257 |
+
"step": 850
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 2.741935483870968,
|
| 261 |
+
"eval_loss": 0.6951683163642883,
|
| 262 |
+
"eval_runtime": 89.0056,
|
| 263 |
+
"eval_samples_per_second": 213.515,
|
| 264 |
+
"eval_steps_per_second": 4.449,
|
| 265 |
+
"step": 850
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 2.903225806451613,
|
| 269 |
+
"grad_norm": 1.0460095405578613,
|
| 270 |
+
"learning_rate": 0.0001884,
|
| 271 |
+
"loss": 0.7883,
|
| 272 |
+
"step": 900
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 2.903225806451613,
|
| 276 |
+
"eval_loss": 0.6605275869369507,
|
| 277 |
+
"eval_runtime": 87.0532,
|
| 278 |
+
"eval_samples_per_second": 218.303,
|
| 279 |
+
"eval_steps_per_second": 4.549,
|
| 280 |
+
"step": 900
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 3.064516129032258,
|
| 284 |
+
"grad_norm": 0.7760242223739624,
|
| 285 |
+
"learning_rate": 0.00018775483870967744,
|
| 286 |
+
"loss": 0.7428,
|
| 287 |
+
"step": 950
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 3.064516129032258,
|
| 291 |
+
"eval_loss": 0.6210550665855408,
|
| 292 |
+
"eval_runtime": 87.5365,
|
| 293 |
+
"eval_samples_per_second": 217.098,
|
| 294 |
+
"eval_steps_per_second": 4.524,
|
| 295 |
+
"step": 950
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 3.225806451612903,
|
| 299 |
+
"grad_norm": 0.7606909275054932,
|
| 300 |
+
"learning_rate": 0.00018710967741935485,
|
| 301 |
+
"loss": 0.7159,
|
| 302 |
+
"step": 1000
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.225806451612903,
|
| 306 |
+
"eval_loss": 0.60401451587677,
|
| 307 |
+
"eval_runtime": 86.4168,
|
| 308 |
+
"eval_samples_per_second": 219.911,
|
| 309 |
+
"eval_steps_per_second": 4.582,
|
| 310 |
+
"step": 1000
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 3.3870967741935485,
|
| 314 |
+
"grad_norm": 0.7674448490142822,
|
| 315 |
+
"learning_rate": 0.00018646451612903226,
|
| 316 |
+
"loss": 0.6965,
|
| 317 |
+
"step": 1050
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 3.3870967741935485,
|
| 321 |
+
"eval_loss": 0.583739697933197,
|
| 322 |
+
"eval_runtime": 87.6612,
|
| 323 |
+
"eval_samples_per_second": 216.789,
|
| 324 |
+
"eval_steps_per_second": 4.517,
|
| 325 |
+
"step": 1050
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 3.5483870967741935,
|
| 329 |
+
"grad_norm": 0.8499358296394348,
|
| 330 |
+
"learning_rate": 0.0001858193548387097,
|
| 331 |
+
"loss": 0.6706,
|
| 332 |
+
"step": 1100
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 3.5483870967741935,
|
| 336 |
+
"eval_loss": 0.567313551902771,
|
| 337 |
+
"eval_runtime": 84.0187,
|
| 338 |
+
"eval_samples_per_second": 226.188,
|
| 339 |
+
"eval_steps_per_second": 4.713,
|
| 340 |
+
"step": 1100
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 3.709677419354839,
|
| 344 |
+
"grad_norm": 0.8062635660171509,
|
| 345 |
+
"learning_rate": 0.0001851741935483871,
|
| 346 |
+
"loss": 0.6605,
|
| 347 |
+
"step": 1150
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 3.709677419354839,
|
| 351 |
+
"eval_loss": 0.5546169281005859,
|
| 352 |
+
"eval_runtime": 87.6004,
|
| 353 |
+
"eval_samples_per_second": 216.94,
|
| 354 |
+
"eval_steps_per_second": 4.521,
|
| 355 |
+
"step": 1150
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 3.870967741935484,
|
| 359 |
+
"grad_norm": 0.9007663726806641,
|
| 360 |
+
"learning_rate": 0.00018452903225806453,
|
| 361 |
+
"loss": 0.6528,
|
| 362 |
+
"step": 1200
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 3.870967741935484,
|
| 366 |
+
"eval_loss": 0.5342397093772888,
|
| 367 |
+
"eval_runtime": 90.2247,
|
| 368 |
+
"eval_samples_per_second": 210.63,
|
| 369 |
+
"eval_steps_per_second": 4.389,
|
| 370 |
+
"step": 1200
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 4.032258064516129,
|
| 374 |
+
"grad_norm": 0.6880891919136047,
|
| 375 |
+
"learning_rate": 0.00018388387096774194,
|
| 376 |
+
"loss": 0.6174,
|
| 377 |
+
"step": 1250
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 4.032258064516129,
|
| 381 |
+
"eval_loss": 0.5208889245986938,
|
| 382 |
+
"eval_runtime": 94.9972,
|
| 383 |
+
"eval_samples_per_second": 200.048,
|
| 384 |
+
"eval_steps_per_second": 4.169,
|
| 385 |
+
"step": 1250
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 4.193548387096774,
|
| 389 |
+
"grad_norm": 0.708258330821991,
|
| 390 |
+
"learning_rate": 0.00018323870967741935,
|
| 391 |
+
"loss": 0.5884,
|
| 392 |
+
"step": 1300
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 4.193548387096774,
|
| 396 |
+
"eval_loss": 0.5109750032424927,
|
| 397 |
+
"eval_runtime": 94.4447,
|
| 398 |
+
"eval_samples_per_second": 201.218,
|
| 399 |
+
"eval_steps_per_second": 4.193,
|
| 400 |
+
"step": 1300
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 4.354838709677419,
|
| 404 |
+
"grad_norm": 0.5771231651306152,
|
| 405 |
+
"learning_rate": 0.0001825935483870968,
|
| 406 |
+
"loss": 0.5515,
|
| 407 |
+
"step": 1350
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 4.354838709677419,
|
| 411 |
+
"eval_loss": 0.5005462169647217,
|
| 412 |
+
"eval_runtime": 92.4886,
|
| 413 |
+
"eval_samples_per_second": 205.474,
|
| 414 |
+
"eval_steps_per_second": 4.282,
|
| 415 |
+
"step": 1350
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 4.516129032258064,
|
| 419 |
+
"grad_norm": 0.9058783054351807,
|
| 420 |
+
"learning_rate": 0.0001819483870967742,
|
| 421 |
+
"loss": 0.5876,
|
| 422 |
+
"step": 1400
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 4.516129032258064,
|
| 426 |
+
"eval_loss": 0.4874018728733063,
|
| 427 |
+
"eval_runtime": 90.1408,
|
| 428 |
+
"eval_samples_per_second": 210.826,
|
| 429 |
+
"eval_steps_per_second": 4.393,
|
| 430 |
+
"step": 1400
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 4.67741935483871,
|
| 434 |
+
"grad_norm": 0.702014148235321,
|
| 435 |
+
"learning_rate": 0.00018130322580645162,
|
| 436 |
+
"loss": 0.5564,
|
| 437 |
+
"step": 1450
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 4.67741935483871,
|
| 441 |
+
"eval_loss": 0.47600802779197693,
|
| 442 |
+
"eval_runtime": 89.1009,
|
| 443 |
+
"eval_samples_per_second": 213.286,
|
| 444 |
+
"eval_steps_per_second": 4.444,
|
| 445 |
+
"step": 1450
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 4.838709677419355,
|
| 449 |
+
"grad_norm": 0.6843072772026062,
|
| 450 |
+
"learning_rate": 0.00018065806451612903,
|
| 451 |
+
"loss": 0.5857,
|
| 452 |
+
"step": 1500
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 4.838709677419355,
|
| 456 |
+
"eval_loss": 0.45552995800971985,
|
| 457 |
+
"eval_runtime": 87.6377,
|
| 458 |
+
"eval_samples_per_second": 216.847,
|
| 459 |
+
"eval_steps_per_second": 4.519,
|
| 460 |
+
"step": 1500
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 5.0,
|
| 464 |
+
"grad_norm": 0.9617322683334351,
|
| 465 |
+
"learning_rate": 0.00018001290322580647,
|
| 466 |
+
"loss": 0.5225,
|
| 467 |
+
"step": 1550
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 5.0,
|
| 471 |
+
"eval_loss": 0.448142945766449,
|
| 472 |
+
"eval_runtime": 88.2995,
|
| 473 |
+
"eval_samples_per_second": 215.222,
|
| 474 |
+
"eval_steps_per_second": 4.485,
|
| 475 |
+
"step": 1550
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 5.161290322580645,
|
| 479 |
+
"grad_norm": 0.5962368249893188,
|
| 480 |
+
"learning_rate": 0.00017936774193548388,
|
| 481 |
+
"loss": 0.5237,
|
| 482 |
+
"step": 1600
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 5.161290322580645,
|
| 486 |
+
"eval_loss": 0.4412171542644501,
|
| 487 |
+
"eval_runtime": 86.7486,
|
| 488 |
+
"eval_samples_per_second": 219.07,
|
| 489 |
+
"eval_steps_per_second": 4.565,
|
| 490 |
+
"step": 1600
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 5.32258064516129,
|
| 494 |
+
"grad_norm": 0.6392622590065002,
|
| 495 |
+
"learning_rate": 0.0001787225806451613,
|
| 496 |
+
"loss": 0.5279,
|
| 497 |
+
"step": 1650
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 5.32258064516129,
|
| 501 |
+
"eval_loss": 0.43270382285118103,
|
| 502 |
+
"eval_runtime": 86.3243,
|
| 503 |
+
"eval_samples_per_second": 220.147,
|
| 504 |
+
"eval_steps_per_second": 4.587,
|
| 505 |
+
"step": 1650
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 5.483870967741936,
|
| 509 |
+
"grad_norm": 0.5868324637413025,
|
| 510 |
+
"learning_rate": 0.00017807741935483873,
|
| 511 |
+
"loss": 0.4941,
|
| 512 |
+
"step": 1700
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 5.483870967741936,
|
| 516 |
+
"eval_loss": 0.42134207487106323,
|
| 517 |
+
"eval_runtime": 87.3073,
|
| 518 |
+
"eval_samples_per_second": 217.668,
|
| 519 |
+
"eval_steps_per_second": 4.536,
|
| 520 |
+
"step": 1700
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 5.645161290322581,
|
| 524 |
+
"grad_norm": 0.6335242986679077,
|
| 525 |
+
"learning_rate": 0.00017743225806451615,
|
| 526 |
+
"loss": 0.4771,
|
| 527 |
+
"step": 1750
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 5.645161290322581,
|
| 531 |
+
"eval_loss": 0.4060479998588562,
|
| 532 |
+
"eval_runtime": 86.2245,
|
| 533 |
+
"eval_samples_per_second": 220.402,
|
| 534 |
+
"eval_steps_per_second": 4.593,
|
| 535 |
+
"step": 1750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 5.806451612903226,
|
| 539 |
+
"grad_norm": 0.6034550666809082,
|
| 540 |
+
"learning_rate": 0.00017678709677419356,
|
| 541 |
+
"loss": 0.4411,
|
| 542 |
+
"step": 1800
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 5.806451612903226,
|
| 546 |
+
"eval_loss": 0.39572522044181824,
|
| 547 |
+
"eval_runtime": 86.9597,
|
| 548 |
+
"eval_samples_per_second": 218.538,
|
| 549 |
+
"eval_steps_per_second": 4.554,
|
| 550 |
+
"step": 1800
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 5.967741935483871,
|
| 554 |
+
"grad_norm": 0.6557337045669556,
|
| 555 |
+
"learning_rate": 0.00017614193548387097,
|
| 556 |
+
"loss": 0.4511,
|
| 557 |
+
"step": 1850
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 5.967741935483871,
|
| 561 |
+
"eval_loss": 0.3814023435115814,
|
| 562 |
+
"eval_runtime": 88.9636,
|
| 563 |
+
"eval_samples_per_second": 213.615,
|
| 564 |
+
"eval_steps_per_second": 4.451,
|
| 565 |
+
"step": 1850
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 6.129032258064516,
|
| 569 |
+
"grad_norm": 0.510995090007782,
|
| 570 |
+
"learning_rate": 0.0001754967741935484,
|
| 571 |
+
"loss": 0.4288,
|
| 572 |
+
"step": 1900
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 6.129032258064516,
|
| 576 |
+
"eval_loss": 0.3621025085449219,
|
| 577 |
+
"eval_runtime": 92.0174,
|
| 578 |
+
"eval_samples_per_second": 206.526,
|
| 579 |
+
"eval_steps_per_second": 4.304,
|
| 580 |
+
"step": 1900
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 6.290322580645161,
|
| 584 |
+
"grad_norm": 0.644413948059082,
|
| 585 |
+
"learning_rate": 0.00017485161290322582,
|
| 586 |
+
"loss": 0.4234,
|
| 587 |
+
"step": 1950
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 6.290322580645161,
|
| 591 |
+
"eval_loss": 0.35021910071372986,
|
| 592 |
+
"eval_runtime": 95.7807,
|
| 593 |
+
"eval_samples_per_second": 198.411,
|
| 594 |
+
"eval_steps_per_second": 4.134,
|
| 595 |
+
"step": 1950
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 6.451612903225806,
|
| 599 |
+
"grad_norm": 0.5785158276557922,
|
| 600 |
+
"learning_rate": 0.00017420645161290323,
|
| 601 |
+
"loss": 0.4009,
|
| 602 |
+
"step": 2000
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 6.451612903225806,
|
| 606 |
+
"eval_loss": 0.33460476994514465,
|
| 607 |
+
"eval_runtime": 146.86,
|
| 608 |
+
"eval_samples_per_second": 129.402,
|
| 609 |
+
"eval_steps_per_second": 2.696,
|
| 610 |
+
"step": 2000
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 6.612903225806452,
|
| 614 |
+
"grad_norm": 0.7037348747253418,
|
| 615 |
+
"learning_rate": 0.00017356129032258067,
|
| 616 |
+
"loss": 0.3922,
|
| 617 |
+
"step": 2050
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 6.612903225806452,
|
| 621 |
+
"eval_loss": 0.3163394331932068,
|
| 622 |
+
"eval_runtime": 179.0958,
|
| 623 |
+
"eval_samples_per_second": 106.111,
|
| 624 |
+
"eval_steps_per_second": 2.211,
|
| 625 |
+
"step": 2050
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 6.774193548387097,
|
| 629 |
+
"grad_norm": 0.6514049768447876,
|
| 630 |
+
"learning_rate": 0.00017291612903225806,
|
| 631 |
+
"loss": 0.3524,
|
| 632 |
+
"step": 2100
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 6.774193548387097,
|
| 636 |
+
"eval_loss": 0.30841001868247986,
|
| 637 |
+
"eval_runtime": 179.6968,
|
| 638 |
+
"eval_samples_per_second": 105.756,
|
| 639 |
+
"eval_steps_per_second": 2.204,
|
| 640 |
+
"step": 2100
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 6.935483870967742,
|
| 644 |
+
"grad_norm": 0.6443042755126953,
|
| 645 |
+
"learning_rate": 0.0001722709677419355,
|
| 646 |
+
"loss": 0.3346,
|
| 647 |
+
"step": 2150
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 6.935483870967742,
|
| 651 |
+
"eval_loss": 0.2951297163963318,
|
| 652 |
+
"eval_runtime": 180.7736,
|
| 653 |
+
"eval_samples_per_second": 105.126,
|
| 654 |
+
"eval_steps_per_second": 2.191,
|
| 655 |
+
"step": 2150
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 7.096774193548387,
|
| 659 |
+
"grad_norm": 0.6390765309333801,
|
| 660 |
+
"learning_rate": 0.0001716258064516129,
|
| 661 |
+
"loss": 0.3255,
|
| 662 |
+
"step": 2200
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 7.096774193548387,
|
| 666 |
+
"eval_loss": 0.2864570915699005,
|
| 667 |
+
"eval_runtime": 177.5655,
|
| 668 |
+
"eval_samples_per_second": 107.025,
|
| 669 |
+
"eval_steps_per_second": 2.23,
|
| 670 |
+
"step": 2200
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 7.258064516129032,
|
| 674 |
+
"grad_norm": 0.6152281165122986,
|
| 675 |
+
"learning_rate": 0.00017098064516129032,
|
| 676 |
+
"loss": 0.3258,
|
| 677 |
+
"step": 2250
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 7.258064516129032,
|
| 681 |
+
"eval_loss": 0.26722875237464905,
|
| 682 |
+
"eval_runtime": 162.1355,
|
| 683 |
+
"eval_samples_per_second": 117.211,
|
| 684 |
+
"eval_steps_per_second": 2.442,
|
| 685 |
+
"step": 2250
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 7.419354838709677,
|
| 689 |
+
"grad_norm": 0.5493067502975464,
|
| 690 |
+
"learning_rate": 0.00017033548387096776,
|
| 691 |
+
"loss": 0.3071,
|
| 692 |
+
"step": 2300
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 7.419354838709677,
|
| 696 |
+
"eval_loss": 0.2526390552520752,
|
| 697 |
+
"eval_runtime": 200.1286,
|
| 698 |
+
"eval_samples_per_second": 94.959,
|
| 699 |
+
"eval_steps_per_second": 1.979,
|
| 700 |
+
"step": 2300
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 7.580645161290323,
|
| 704 |
+
"grad_norm": 0.5964747667312622,
|
| 705 |
+
"learning_rate": 0.00016969032258064518,
|
| 706 |
+
"loss": 0.288,
|
| 707 |
+
"step": 2350
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 7.580645161290323,
|
| 711 |
+
"eval_loss": 0.24554206430912018,
|
| 712 |
+
"eval_runtime": 275.0797,
|
| 713 |
+
"eval_samples_per_second": 69.085,
|
| 714 |
+
"eval_steps_per_second": 1.44,
|
| 715 |
+
"step": 2350
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 7.741935483870968,
|
| 719 |
+
"grad_norm": 0.5316987037658691,
|
| 720 |
+
"learning_rate": 0.0001690451612903226,
|
| 721 |
+
"loss": 0.2739,
|
| 722 |
+
"step": 2400
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 7.741935483870968,
|
| 726 |
+
"eval_loss": 0.23691914975643158,
|
| 727 |
+
"eval_runtime": 175.1655,
|
| 728 |
+
"eval_samples_per_second": 108.492,
|
| 729 |
+
"eval_steps_per_second": 2.261,
|
| 730 |
+
"step": 2400
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 7.903225806451613,
|
| 734 |
+
"grad_norm": 0.47612282633781433,
|
| 735 |
+
"learning_rate": 0.0001684,
|
| 736 |
+
"loss": 0.2706,
|
| 737 |
+
"step": 2450
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 7.903225806451613,
|
| 741 |
+
"eval_loss": 0.23088718950748444,
|
| 742 |
+
"eval_runtime": 174.1382,
|
| 743 |
+
"eval_samples_per_second": 109.132,
|
| 744 |
+
"eval_steps_per_second": 2.274,
|
| 745 |
+
"step": 2450
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 8.064516129032258,
|
| 749 |
+
"grad_norm": 0.5631929039955139,
|
| 750 |
+
"learning_rate": 0.00016775483870967744,
|
| 751 |
+
"loss": 0.2645,
|
| 752 |
+
"step": 2500
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 8.064516129032258,
|
| 756 |
+
"eval_loss": 0.229031041264534,
|
| 757 |
+
"eval_runtime": 321.9761,
|
| 758 |
+
"eval_samples_per_second": 59.023,
|
| 759 |
+
"eval_steps_per_second": 1.23,
|
| 760 |
+
"step": 2500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 8.225806451612904,
|
| 764 |
+
"grad_norm": 0.5553293228149414,
|
| 765 |
+
"learning_rate": 0.00016710967741935483,
|
| 766 |
+
"loss": 0.2604,
|
| 767 |
+
"step": 2550
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 8.225806451612904,
|
| 771 |
+
"eval_loss": 0.22551080584526062,
|
| 772 |
+
"eval_runtime": 388.7037,
|
| 773 |
+
"eval_samples_per_second": 48.891,
|
| 774 |
+
"eval_steps_per_second": 1.019,
|
| 775 |
+
"step": 2550
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 8.387096774193548,
|
| 779 |
+
"grad_norm": 0.5779751539230347,
|
| 780 |
+
"learning_rate": 0.00016646451612903226,
|
| 781 |
+
"loss": 0.2537,
|
| 782 |
+
"step": 2600
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 8.387096774193548,
|
| 786 |
+
"eval_loss": 0.22034035623073578,
|
| 787 |
+
"eval_runtime": 296.8732,
|
| 788 |
+
"eval_samples_per_second": 64.014,
|
| 789 |
+
"eval_steps_per_second": 1.334,
|
| 790 |
+
"step": 2600
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 8.548387096774194,
|
| 794 |
+
"grad_norm": 0.4632438123226166,
|
| 795 |
+
"learning_rate": 0.0001658193548387097,
|
| 796 |
+
"loss": 0.2505,
|
| 797 |
+
"step": 2650
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 8.548387096774194,
|
| 801 |
+
"eval_loss": 0.21870557963848114,
|
| 802 |
+
"eval_runtime": 195.9343,
|
| 803 |
+
"eval_samples_per_second": 96.992,
|
| 804 |
+
"eval_steps_per_second": 2.021,
|
| 805 |
+
"step": 2650
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 8.709677419354838,
|
| 809 |
+
"grad_norm": 0.5242488980293274,
|
| 810 |
+
"learning_rate": 0.0001651741935483871,
|
| 811 |
+
"loss": 0.2465,
|
| 812 |
+
"step": 2700
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 8.709677419354838,
|
| 816 |
+
"eval_loss": 0.21318656206130981,
|
| 817 |
+
"eval_runtime": 199.2062,
|
| 818 |
+
"eval_samples_per_second": 95.399,
|
| 819 |
+
"eval_steps_per_second": 1.988,
|
| 820 |
+
"step": 2700
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 8.870967741935484,
|
| 824 |
+
"grad_norm": 0.5110979676246643,
|
| 825 |
+
"learning_rate": 0.00016452903225806453,
|
| 826 |
+
"loss": 0.2411,
|
| 827 |
+
"step": 2750
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 8.870967741935484,
|
| 831 |
+
"eval_loss": 0.21142736077308655,
|
| 832 |
+
"eval_runtime": 196.2975,
|
| 833 |
+
"eval_samples_per_second": 96.812,
|
| 834 |
+
"eval_steps_per_second": 2.017,
|
| 835 |
+
"step": 2750
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 9.03225806451613,
|
| 839 |
+
"grad_norm": 0.4415765702724457,
|
| 840 |
+
"learning_rate": 0.00016388387096774194,
|
| 841 |
+
"loss": 0.2431,
|
| 842 |
+
"step": 2800
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 9.03225806451613,
|
| 846 |
+
"eval_loss": 0.20968259871006012,
|
| 847 |
+
"eval_runtime": 190.8363,
|
| 848 |
+
"eval_samples_per_second": 99.583,
|
| 849 |
+
"eval_steps_per_second": 2.075,
|
| 850 |
+
"step": 2800
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 9.193548387096774,
|
| 854 |
+
"grad_norm": 0.4640558362007141,
|
| 855 |
+
"learning_rate": 0.00016323870967741935,
|
| 856 |
+
"loss": 0.2416,
|
| 857 |
+
"step": 2850
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 9.193548387096774,
|
| 861 |
+
"eval_loss": 0.206209197640419,
|
| 862 |
+
"eval_runtime": 219.8043,
|
| 863 |
+
"eval_samples_per_second": 86.459,
|
| 864 |
+
"eval_steps_per_second": 1.802,
|
| 865 |
+
"step": 2850
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 9.35483870967742,
|
| 869 |
+
"grad_norm": 0.5190003514289856,
|
| 870 |
+
"learning_rate": 0.00016259354838709677,
|
| 871 |
+
"loss": 0.2321,
|
| 872 |
+
"step": 2900
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 9.35483870967742,
|
| 876 |
+
"eval_loss": 0.20502887666225433,
|
| 877 |
+
"eval_runtime": 232.185,
|
| 878 |
+
"eval_samples_per_second": 81.849,
|
| 879 |
+
"eval_steps_per_second": 1.706,
|
| 880 |
+
"step": 2900
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 9.516129032258064,
|
| 884 |
+
"grad_norm": 0.5595571398735046,
|
| 885 |
+
"learning_rate": 0.0001619483870967742,
|
| 886 |
+
"loss": 0.2317,
|
| 887 |
+
"step": 2950
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 9.516129032258064,
|
| 891 |
+
"eval_loss": 0.20400060713291168,
|
| 892 |
+
"eval_runtime": 295.1096,
|
| 893 |
+
"eval_samples_per_second": 64.396,
|
| 894 |
+
"eval_steps_per_second": 1.342,
|
| 895 |
+
"step": 2950
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 9.67741935483871,
|
| 899 |
+
"grad_norm": 0.5060888528823853,
|
| 900 |
+
"learning_rate": 0.00016130322580645162,
|
| 901 |
+
"loss": 0.2263,
|
| 902 |
+
"step": 3000
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 9.67741935483871,
|
| 906 |
+
"eval_loss": 0.2005593329668045,
|
| 907 |
+
"eval_runtime": 201.7676,
|
| 908 |
+
"eval_samples_per_second": 94.188,
|
| 909 |
+
"eval_steps_per_second": 1.963,
|
| 910 |
+
"step": 3000
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 9.838709677419354,
|
| 914 |
+
"grad_norm": 0.5235562920570374,
|
| 915 |
+
"learning_rate": 0.00016065806451612903,
|
| 916 |
+
"loss": 0.2209,
|
| 917 |
+
"step": 3050
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 9.838709677419354,
|
| 921 |
+
"eval_loss": 0.2015007883310318,
|
| 922 |
+
"eval_runtime": 200.5101,
|
| 923 |
+
"eval_samples_per_second": 94.778,
|
| 924 |
+
"eval_steps_per_second": 1.975,
|
| 925 |
+
"step": 3050
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 10.0,
|
| 929 |
+
"grad_norm": 0.7316103577613831,
|
| 930 |
+
"learning_rate": 0.00016001290322580647,
|
| 931 |
+
"loss": 0.2184,
|
| 932 |
+
"step": 3100
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 10.0,
|
| 936 |
+
"eval_loss": 0.1977168172597885,
|
| 937 |
+
"eval_runtime": 185.6624,
|
| 938 |
+
"eval_samples_per_second": 102.358,
|
| 939 |
+
"eval_steps_per_second": 2.133,
|
| 940 |
+
"step": 3100
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 10.161290322580646,
|
| 944 |
+
"grad_norm": 0.4775584638118744,
|
| 945 |
+
"learning_rate": 0.00015936774193548388,
|
| 946 |
+
"loss": 0.2204,
|
| 947 |
+
"step": 3150
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 10.161290322580646,
|
| 951 |
+
"eval_loss": 0.19870446622371674,
|
| 952 |
+
"eval_runtime": 192.3529,
|
| 953 |
+
"eval_samples_per_second": 98.798,
|
| 954 |
+
"eval_steps_per_second": 2.059,
|
| 955 |
+
"step": 3150
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 10.32258064516129,
|
| 959 |
+
"grad_norm": 0.4865228533744812,
|
| 960 |
+
"learning_rate": 0.0001587225806451613,
|
| 961 |
+
"loss": 0.2152,
|
| 962 |
+
"step": 3200
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 10.32258064516129,
|
| 966 |
+
"eval_loss": 0.19557693600654602,
|
| 967 |
+
"eval_runtime": 195.6795,
|
| 968 |
+
"eval_samples_per_second": 97.118,
|
| 969 |
+
"eval_steps_per_second": 2.024,
|
| 970 |
+
"step": 3200
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 10.483870967741936,
|
| 974 |
+
"grad_norm": 0.5472003221511841,
|
| 975 |
+
"learning_rate": 0.0001580774193548387,
|
| 976 |
+
"loss": 0.2188,
|
| 977 |
+
"step": 3250
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 10.483870967741936,
|
| 981 |
+
"eval_loss": 0.19073671102523804,
|
| 982 |
+
"eval_runtime": 188.5485,
|
| 983 |
+
"eval_samples_per_second": 100.791,
|
| 984 |
+
"eval_steps_per_second": 2.1,
|
| 985 |
+
"step": 3250
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 10.64516129032258,
|
| 989 |
+
"grad_norm": 0.46814078092575073,
|
| 990 |
+
"learning_rate": 0.00015743225806451615,
|
| 991 |
+
"loss": 0.2122,
|
| 992 |
+
"step": 3300
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 10.64516129032258,
|
| 996 |
+
"eval_loss": 0.1892285943031311,
|
| 997 |
+
"eval_runtime": 190.3556,
|
| 998 |
+
"eval_samples_per_second": 99.834,
|
| 999 |
+
"eval_steps_per_second": 2.08,
|
| 1000 |
+
"step": 3300
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 10.806451612903226,
|
| 1004 |
+
"grad_norm": 0.39085471630096436,
|
| 1005 |
+
"learning_rate": 0.00015678709677419356,
|
| 1006 |
+
"loss": 0.2157,
|
| 1007 |
+
"step": 3350
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 10.806451612903226,
|
| 1011 |
+
"eval_loss": 0.19001252949237823,
|
| 1012 |
+
"eval_runtime": 189.4897,
|
| 1013 |
+
"eval_samples_per_second": 100.29,
|
| 1014 |
+
"eval_steps_per_second": 2.09,
|
| 1015 |
+
"step": 3350
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 10.967741935483872,
|
| 1019 |
+
"grad_norm": 0.4231501519680023,
|
| 1020 |
+
"learning_rate": 0.00015614193548387097,
|
| 1021 |
+
"loss": 0.2162,
|
| 1022 |
+
"step": 3400
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 10.967741935483872,
|
| 1026 |
+
"eval_loss": 0.18620692193508148,
|
| 1027 |
+
"eval_runtime": 190.5331,
|
| 1028 |
+
"eval_samples_per_second": 99.741,
|
| 1029 |
+
"eval_steps_per_second": 2.078,
|
| 1030 |
+
"step": 3400
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 11.129032258064516,
|
| 1034 |
+
"grad_norm": 0.5186159610748291,
|
| 1035 |
+
"learning_rate": 0.0001554967741935484,
|
| 1036 |
+
"loss": 0.2101,
|
| 1037 |
+
"step": 3450
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 11.129032258064516,
|
| 1041 |
+
"eval_loss": 0.18647471070289612,
|
| 1042 |
+
"eval_runtime": 188.7317,
|
| 1043 |
+
"eval_samples_per_second": 100.693,
|
| 1044 |
+
"eval_steps_per_second": 2.098,
|
| 1045 |
+
"step": 3450
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 11.290322580645162,
|
| 1049 |
+
"grad_norm": 0.4359528124332428,
|
| 1050 |
+
"learning_rate": 0.0001548516129032258,
|
| 1051 |
+
"loss": 0.2051,
|
| 1052 |
+
"step": 3500
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 11.290322580645162,
|
| 1056 |
+
"eval_loss": 0.18508924543857574,
|
| 1057 |
+
"eval_runtime": 187.5602,
|
| 1058 |
+
"eval_samples_per_second": 101.322,
|
| 1059 |
+
"eval_steps_per_second": 2.111,
|
| 1060 |
+
"step": 3500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 11.451612903225806,
|
| 1064 |
+
"grad_norm": 0.35816341638565063,
|
| 1065 |
+
"learning_rate": 0.00015420645161290324,
|
| 1066 |
+
"loss": 0.2027,
|
| 1067 |
+
"step": 3550
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 11.451612903225806,
|
| 1071 |
+
"eval_loss": 0.18324784934520721,
|
| 1072 |
+
"eval_runtime": 188.342,
|
| 1073 |
+
"eval_samples_per_second": 100.902,
|
| 1074 |
+
"eval_steps_per_second": 2.103,
|
| 1075 |
+
"step": 3550
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 11.612903225806452,
|
| 1079 |
+
"grad_norm": 0.4818542003631592,
|
| 1080 |
+
"learning_rate": 0.00015356129032258065,
|
| 1081 |
+
"loss": 0.2059,
|
| 1082 |
+
"step": 3600
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 11.612903225806452,
|
| 1086 |
+
"eval_loss": 0.18456551432609558,
|
| 1087 |
+
"eval_runtime": 190.2916,
|
| 1088 |
+
"eval_samples_per_second": 99.868,
|
| 1089 |
+
"eval_steps_per_second": 2.081,
|
| 1090 |
+
"step": 3600
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 11.774193548387096,
|
| 1094 |
+
"grad_norm": 0.560644805431366,
|
| 1095 |
+
"learning_rate": 0.00015291612903225806,
|
| 1096 |
+
"loss": 0.1994,
|
| 1097 |
+
"step": 3650
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 11.774193548387096,
|
| 1101 |
+
"eval_loss": 0.1837019920349121,
|
| 1102 |
+
"eval_runtime": 193.7316,
|
| 1103 |
+
"eval_samples_per_second": 98.094,
|
| 1104 |
+
"eval_steps_per_second": 2.044,
|
| 1105 |
+
"step": 3650
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 11.935483870967742,
|
| 1109 |
+
"grad_norm": 0.4187397360801697,
|
| 1110 |
+
"learning_rate": 0.0001522709677419355,
|
| 1111 |
+
"loss": 0.208,
|
| 1112 |
+
"step": 3700
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 11.935483870967742,
|
| 1116 |
+
"eval_loss": 0.18199948966503143,
|
| 1117 |
+
"eval_runtime": 190.9935,
|
| 1118 |
+
"eval_samples_per_second": 99.501,
|
| 1119 |
+
"eval_steps_per_second": 2.073,
|
| 1120 |
+
"step": 3700
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 12.096774193548388,
|
| 1124 |
+
"grad_norm": 0.47370871901512146,
|
| 1125 |
+
"learning_rate": 0.0001516258064516129,
|
| 1126 |
+
"loss": 0.197,
|
| 1127 |
+
"step": 3750
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 12.096774193548388,
|
| 1131 |
+
"eval_loss": 0.18183062970638275,
|
| 1132 |
+
"eval_runtime": 189.271,
|
| 1133 |
+
"eval_samples_per_second": 100.406,
|
| 1134 |
+
"eval_steps_per_second": 2.092,
|
| 1135 |
+
"step": 3750
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 12.258064516129032,
|
| 1139 |
+
"grad_norm": 0.5040688514709473,
|
| 1140 |
+
"learning_rate": 0.00015098064516129033,
|
| 1141 |
+
"loss": 0.1984,
|
| 1142 |
+
"step": 3800
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 12.258064516129032,
|
| 1146 |
+
"eval_loss": 0.17957282066345215,
|
| 1147 |
+
"eval_runtime": 187.7702,
|
| 1148 |
+
"eval_samples_per_second": 101.209,
|
| 1149 |
+
"eval_steps_per_second": 2.109,
|
| 1150 |
+
"step": 3800
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 12.419354838709678,
|
| 1154 |
+
"grad_norm": 0.44129958748817444,
|
| 1155 |
+
"learning_rate": 0.00015033548387096774,
|
| 1156 |
+
"loss": 0.1905,
|
| 1157 |
+
"step": 3850
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 12.419354838709678,
|
| 1161 |
+
"eval_loss": 0.17632746696472168,
|
| 1162 |
+
"eval_runtime": 189.2669,
|
| 1163 |
+
"eval_samples_per_second": 100.408,
|
| 1164 |
+
"eval_steps_per_second": 2.092,
|
| 1165 |
+
"step": 3850
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 12.580645161290322,
|
| 1169 |
+
"grad_norm": 0.38648995757102966,
|
| 1170 |
+
"learning_rate": 0.00014969032258064518,
|
| 1171 |
+
"loss": 0.1915,
|
| 1172 |
+
"step": 3900
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 12.580645161290322,
|
| 1176 |
+
"eval_loss": 0.1763821244239807,
|
| 1177 |
+
"eval_runtime": 187.8282,
|
| 1178 |
+
"eval_samples_per_second": 101.178,
|
| 1179 |
+
"eval_steps_per_second": 2.108,
|
| 1180 |
+
"step": 3900
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 12.741935483870968,
|
| 1184 |
+
"grad_norm": 0.46786022186279297,
|
| 1185 |
+
"learning_rate": 0.0001490451612903226,
|
| 1186 |
+
"loss": 0.1964,
|
| 1187 |
+
"step": 3950
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 12.741935483870968,
|
| 1191 |
+
"eval_loss": 0.1770430952310562,
|
| 1192 |
+
"eval_runtime": 192.9563,
|
| 1193 |
+
"eval_samples_per_second": 98.489,
|
| 1194 |
+
"eval_steps_per_second": 2.052,
|
| 1195 |
+
"step": 3950
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 12.903225806451612,
|
| 1199 |
+
"grad_norm": 0.45657825469970703,
|
| 1200 |
+
"learning_rate": 0.0001484,
|
| 1201 |
+
"loss": 0.194,
|
| 1202 |
+
"step": 4000
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 12.903225806451612,
|
| 1206 |
+
"eval_loss": 0.1744171380996704,
|
| 1207 |
+
"eval_runtime": 191.9963,
|
| 1208 |
+
"eval_samples_per_second": 98.981,
|
| 1209 |
+
"eval_steps_per_second": 2.063,
|
| 1210 |
+
"step": 4000
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 13.064516129032258,
|
| 1214 |
+
"grad_norm": 0.4950830936431885,
|
| 1215 |
+
"learning_rate": 0.00014775483870967744,
|
| 1216 |
+
"loss": 0.1897,
|
| 1217 |
+
"step": 4050
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 13.064516129032258,
|
| 1221 |
+
"eval_loss": 0.1715860813856125,
|
| 1222 |
+
"eval_runtime": 190.2705,
|
| 1223 |
+
"eval_samples_per_second": 99.879,
|
| 1224 |
+
"eval_steps_per_second": 2.081,
|
| 1225 |
+
"step": 4050
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 13.225806451612904,
|
| 1229 |
+
"grad_norm": 0.4676801562309265,
|
| 1230 |
+
"learning_rate": 0.00014710967741935485,
|
| 1231 |
+
"loss": 0.1837,
|
| 1232 |
+
"step": 4100
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 13.225806451612904,
|
| 1236 |
+
"eval_loss": 0.16961637139320374,
|
| 1237 |
+
"eval_runtime": 177.9195,
|
| 1238 |
+
"eval_samples_per_second": 106.812,
|
| 1239 |
+
"eval_steps_per_second": 2.226,
|
| 1240 |
+
"step": 4100
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 13.387096774193548,
|
| 1244 |
+
"grad_norm": 0.3973780870437622,
|
| 1245 |
+
"learning_rate": 0.00014646451612903227,
|
| 1246 |
+
"loss": 0.1856,
|
| 1247 |
+
"step": 4150
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 13.387096774193548,
|
| 1251 |
+
"eval_loss": 0.17067177593708038,
|
| 1252 |
+
"eval_runtime": 197.9884,
|
| 1253 |
+
"eval_samples_per_second": 95.985,
|
| 1254 |
+
"eval_steps_per_second": 2.0,
|
| 1255 |
+
"step": 4150
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 13.548387096774194,
|
| 1259 |
+
"grad_norm": 0.46780818700790405,
|
| 1260 |
+
"learning_rate": 0.00014581935483870968,
|
| 1261 |
+
"loss": 0.186,
|
| 1262 |
+
"step": 4200
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 13.548387096774194,
|
| 1266 |
+
"eval_loss": 0.17197734117507935,
|
| 1267 |
+
"eval_runtime": 196.3462,
|
| 1268 |
+
"eval_samples_per_second": 96.788,
|
| 1269 |
+
"eval_steps_per_second": 2.017,
|
| 1270 |
+
"step": 4200
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 13.709677419354838,
|
| 1274 |
+
"grad_norm": 0.5169796347618103,
|
| 1275 |
+
"learning_rate": 0.00014517419354838712,
|
| 1276 |
+
"loss": 0.1886,
|
| 1277 |
+
"step": 4250
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 13.709677419354838,
|
| 1281 |
+
"eval_loss": 0.1707415133714676,
|
| 1282 |
+
"eval_runtime": 194.2514,
|
| 1283 |
+
"eval_samples_per_second": 97.832,
|
| 1284 |
+
"eval_steps_per_second": 2.039,
|
| 1285 |
+
"step": 4250
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 13.870967741935484,
|
| 1289 |
+
"grad_norm": 0.36574116349220276,
|
| 1290 |
+
"learning_rate": 0.00014452903225806453,
|
| 1291 |
+
"loss": 0.1857,
|
| 1292 |
+
"step": 4300
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 13.870967741935484,
|
| 1296 |
+
"eval_loss": 0.16808755695819855,
|
| 1297 |
+
"eval_runtime": 192.206,
|
| 1298 |
+
"eval_samples_per_second": 98.873,
|
| 1299 |
+
"eval_steps_per_second": 2.06,
|
| 1300 |
+
"step": 4300
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 14.03225806451613,
|
| 1304 |
+
"grad_norm": 0.4412609338760376,
|
| 1305 |
+
"learning_rate": 0.00014388387096774194,
|
| 1306 |
+
"loss": 0.1819,
|
| 1307 |
+
"step": 4350
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 14.03225806451613,
|
| 1311 |
+
"eval_loss": 0.17015832662582397,
|
| 1312 |
+
"eval_runtime": 194.1959,
|
| 1313 |
+
"eval_samples_per_second": 97.86,
|
| 1314 |
+
"eval_steps_per_second": 2.039,
|
| 1315 |
+
"step": 4350
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 14.193548387096774,
|
| 1319 |
+
"grad_norm": 0.35714399814605713,
|
| 1320 |
+
"learning_rate": 0.00014323870967741938,
|
| 1321 |
+
"loss": 0.179,
|
| 1322 |
+
"step": 4400
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 14.193548387096774,
|
| 1326 |
+
"eval_loss": 0.16659317910671234,
|
| 1327 |
+
"eval_runtime": 190.7696,
|
| 1328 |
+
"eval_samples_per_second": 99.618,
|
| 1329 |
+
"eval_steps_per_second": 2.076,
|
| 1330 |
+
"step": 4400
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 14.35483870967742,
|
| 1334 |
+
"grad_norm": 0.5498349070549011,
|
| 1335 |
+
"learning_rate": 0.00014259354838709677,
|
| 1336 |
+
"loss": 0.174,
|
| 1337 |
+
"step": 4450
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 14.35483870967742,
|
| 1341 |
+
"eval_loss": 0.166724294424057,
|
| 1342 |
+
"eval_runtime": 194.2466,
|
| 1343 |
+
"eval_samples_per_second": 97.834,
|
| 1344 |
+
"eval_steps_per_second": 2.039,
|
| 1345 |
+
"step": 4450
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 14.516129032258064,
|
| 1349 |
+
"grad_norm": 0.3779986798763275,
|
| 1350 |
+
"learning_rate": 0.0001419483870967742,
|
| 1351 |
+
"loss": 0.18,
|
| 1352 |
+
"step": 4500
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 14.516129032258064,
|
| 1356 |
+
"eval_loss": 0.16606487333774567,
|
| 1357 |
+
"eval_runtime": 193.3423,
|
| 1358 |
+
"eval_samples_per_second": 98.292,
|
| 1359 |
+
"eval_steps_per_second": 2.048,
|
| 1360 |
+
"step": 4500
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 14.67741935483871,
|
| 1364 |
+
"grad_norm": 0.5511322021484375,
|
| 1365 |
+
"learning_rate": 0.00014130322580645162,
|
| 1366 |
+
"loss": 0.1789,
|
| 1367 |
+
"step": 4550
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 14.67741935483871,
|
| 1371 |
+
"eval_loss": 0.16656480729579926,
|
| 1372 |
+
"eval_runtime": 270.6846,
|
| 1373 |
+
"eval_samples_per_second": 70.207,
|
| 1374 |
+
"eval_steps_per_second": 1.463,
|
| 1375 |
+
"step": 4550
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 14.838709677419354,
|
| 1379 |
+
"grad_norm": 0.3705432415008545,
|
| 1380 |
+
"learning_rate": 0.00014065806451612903,
|
| 1381 |
+
"loss": 0.1793,
|
| 1382 |
+
"step": 4600
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 14.838709677419354,
|
| 1386 |
+
"eval_loss": 0.1630878746509552,
|
| 1387 |
+
"eval_runtime": 203.349,
|
| 1388 |
+
"eval_samples_per_second": 93.455,
|
| 1389 |
+
"eval_steps_per_second": 1.947,
|
| 1390 |
+
"step": 4600
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 15.0,
|
| 1394 |
+
"grad_norm": 0.6959958076477051,
|
| 1395 |
+
"learning_rate": 0.00014001290322580647,
|
| 1396 |
+
"loss": 0.1849,
|
| 1397 |
+
"step": 4650
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 15.0,
|
| 1401 |
+
"eval_loss": 0.16222645342350006,
|
| 1402 |
+
"eval_runtime": 196.7729,
|
| 1403 |
+
"eval_samples_per_second": 96.578,
|
| 1404 |
+
"eval_steps_per_second": 2.012,
|
| 1405 |
+
"step": 4650
|
| 1406 |
+
},
|
| 1407 |
+
{
|
| 1408 |
+
"epoch": 15.161290322580646,
|
| 1409 |
+
"grad_norm": 0.38614702224731445,
|
| 1410 |
+
"learning_rate": 0.00013936774193548388,
|
| 1411 |
+
"loss": 0.1729,
|
| 1412 |
+
"step": 4700
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 15.161290322580646,
|
| 1416 |
+
"eval_loss": 0.16200029850006104,
|
| 1417 |
+
"eval_runtime": 191.2671,
|
| 1418 |
+
"eval_samples_per_second": 99.358,
|
| 1419 |
+
"eval_steps_per_second": 2.07,
|
| 1420 |
+
"step": 4700
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 15.32258064516129,
|
| 1424 |
+
"grad_norm": 0.4022436738014221,
|
| 1425 |
+
"learning_rate": 0.0001387225806451613,
|
| 1426 |
+
"loss": 0.1766,
|
| 1427 |
+
"step": 4750
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 15.32258064516129,
|
| 1431 |
+
"eval_loss": 0.1640605926513672,
|
| 1432 |
+
"eval_runtime": 198.7156,
|
| 1433 |
+
"eval_samples_per_second": 95.634,
|
| 1434 |
+
"eval_steps_per_second": 1.993,
|
| 1435 |
+
"step": 4750
|
| 1436 |
+
},
|
| 1437 |
+
{
|
| 1438 |
+
"epoch": 15.483870967741936,
|
| 1439 |
+
"grad_norm": 0.45434874296188354,
|
| 1440 |
+
"learning_rate": 0.0001380774193548387,
|
| 1441 |
+
"loss": 0.1698,
|
| 1442 |
+
"step": 4800
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 15.483870967741936,
|
| 1446 |
+
"eval_loss": 0.16334731876850128,
|
| 1447 |
+
"eval_runtime": 188.1897,
|
| 1448 |
+
"eval_samples_per_second": 100.983,
|
| 1449 |
+
"eval_steps_per_second": 2.104,
|
| 1450 |
+
"step": 4800
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 15.64516129032258,
|
| 1454 |
+
"grad_norm": 0.42634057998657227,
|
| 1455 |
+
"learning_rate": 0.00013743225806451615,
|
| 1456 |
+
"loss": 0.1766,
|
| 1457 |
+
"step": 4850
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 15.64516129032258,
|
| 1461 |
+
"eval_loss": 0.16068318486213684,
|
| 1462 |
+
"eval_runtime": 190.5766,
|
| 1463 |
+
"eval_samples_per_second": 99.718,
|
| 1464 |
+
"eval_steps_per_second": 2.078,
|
| 1465 |
+
"step": 4850
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 15.806451612903226,
|
| 1469 |
+
"grad_norm": 0.504154622554779,
|
| 1470 |
+
"learning_rate": 0.00013678709677419353,
|
| 1471 |
+
"loss": 0.1678,
|
| 1472 |
+
"step": 4900
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 15.806451612903226,
|
| 1476 |
+
"eval_loss": 0.15998931229114532,
|
| 1477 |
+
"eval_runtime": 185.5464,
|
| 1478 |
+
"eval_samples_per_second": 102.422,
|
| 1479 |
+
"eval_steps_per_second": 2.134,
|
| 1480 |
+
"step": 4900
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 15.967741935483872,
|
| 1484 |
+
"grad_norm": 0.5468097925186157,
|
| 1485 |
+
"learning_rate": 0.00013614193548387097,
|
| 1486 |
+
"loss": 0.1792,
|
| 1487 |
+
"step": 4950
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 15.967741935483872,
|
| 1491 |
+
"eval_loss": 0.15972544252872467,
|
| 1492 |
+
"eval_runtime": 186.5203,
|
| 1493 |
+
"eval_samples_per_second": 101.887,
|
| 1494 |
+
"eval_steps_per_second": 2.123,
|
| 1495 |
+
"step": 4950
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 16.129032258064516,
|
| 1499 |
+
"grad_norm": 0.4495028853416443,
|
| 1500 |
+
"learning_rate": 0.0001354967741935484,
|
| 1501 |
+
"loss": 0.1755,
|
| 1502 |
+
"step": 5000
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 16.129032258064516,
|
| 1506 |
+
"eval_loss": 0.1596982628107071,
|
| 1507 |
+
"eval_runtime": 194.6559,
|
| 1508 |
+
"eval_samples_per_second": 97.629,
|
| 1509 |
+
"eval_steps_per_second": 2.034,
|
| 1510 |
+
"step": 5000
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 16.29032258064516,
|
| 1514 |
+
"grad_norm": 0.36081984639167786,
|
| 1515 |
+
"learning_rate": 0.0001348516129032258,
|
| 1516 |
+
"loss": 0.1659,
|
| 1517 |
+
"step": 5050
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 16.29032258064516,
|
| 1521 |
+
"eval_loss": 0.15741322934627533,
|
| 1522 |
+
"eval_runtime": 193.3704,
|
| 1523 |
+
"eval_samples_per_second": 98.278,
|
| 1524 |
+
"eval_steps_per_second": 2.048,
|
| 1525 |
+
"step": 5050
|
| 1526 |
+
},
|
| 1527 |
+
{
|
| 1528 |
+
"epoch": 16.451612903225808,
|
| 1529 |
+
"grad_norm": 0.3472287654876709,
|
| 1530 |
+
"learning_rate": 0.00013420645161290324,
|
| 1531 |
+
"loss": 0.1672,
|
| 1532 |
+
"step": 5100
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 16.451612903225808,
|
| 1536 |
+
"eval_loss": 0.16025249660015106,
|
| 1537 |
+
"eval_runtime": 177.054,
|
| 1538 |
+
"eval_samples_per_second": 107.334,
|
| 1539 |
+
"eval_steps_per_second": 2.237,
|
| 1540 |
+
"step": 5100
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 16.612903225806452,
|
| 1544 |
+
"grad_norm": 0.4033275842666626,
|
| 1545 |
+
"learning_rate": 0.00013356129032258065,
|
| 1546 |
+
"loss": 0.1687,
|
| 1547 |
+
"step": 5150
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 16.612903225806452,
|
| 1551 |
+
"eval_loss": 0.15610146522521973,
|
| 1552 |
+
"eval_runtime": 197.4165,
|
| 1553 |
+
"eval_samples_per_second": 96.263,
|
| 1554 |
+
"eval_steps_per_second": 2.006,
|
| 1555 |
+
"step": 5150
|
| 1556 |
+
},
|
| 1557 |
+
{
|
| 1558 |
+
"epoch": 16.774193548387096,
|
| 1559 |
+
"grad_norm": 0.43590107560157776,
|
| 1560 |
+
"learning_rate": 0.00013291612903225806,
|
| 1561 |
+
"loss": 0.1685,
|
| 1562 |
+
"step": 5200
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 16.774193548387096,
|
| 1566 |
+
"eval_loss": 0.15723493695259094,
|
| 1567 |
+
"eval_runtime": 195.3122,
|
| 1568 |
+
"eval_samples_per_second": 97.301,
|
| 1569 |
+
"eval_steps_per_second": 2.028,
|
| 1570 |
+
"step": 5200
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 16.93548387096774,
|
| 1574 |
+
"grad_norm": 0.33427444100379944,
|
| 1575 |
+
"learning_rate": 0.00013227096774193548,
|
| 1576 |
+
"loss": 0.1711,
|
| 1577 |
+
"step": 5250
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 16.93548387096774,
|
| 1581 |
+
"eval_loss": 0.15542824566364288,
|
| 1582 |
+
"eval_runtime": 192.6293,
|
| 1583 |
+
"eval_samples_per_second": 98.656,
|
| 1584 |
+
"eval_steps_per_second": 2.056,
|
| 1585 |
+
"step": 5250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 17.096774193548388,
|
| 1589 |
+
"grad_norm": 0.38596487045288086,
|
| 1590 |
+
"learning_rate": 0.00013162580645161291,
|
| 1591 |
+
"loss": 0.1662,
|
| 1592 |
+
"step": 5300
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 17.096774193548388,
|
| 1596 |
+
"eval_loss": 0.15477755665779114,
|
| 1597 |
+
"eval_runtime": 218.9351,
|
| 1598 |
+
"eval_samples_per_second": 86.802,
|
| 1599 |
+
"eval_steps_per_second": 1.809,
|
| 1600 |
+
"step": 5300
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 17.258064516129032,
|
| 1604 |
+
"grad_norm": 0.30149680376052856,
|
| 1605 |
+
"learning_rate": 0.00013098064516129033,
|
| 1606 |
+
"loss": 0.1632,
|
| 1607 |
+
"step": 5350
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 17.258064516129032,
|
| 1611 |
+
"eval_loss": 0.15298867225646973,
|
| 1612 |
+
"eval_runtime": 109.5329,
|
| 1613 |
+
"eval_samples_per_second": 173.5,
|
| 1614 |
+
"eval_steps_per_second": 3.615,
|
| 1615 |
+
"step": 5350
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 17.419354838709676,
|
| 1619 |
+
"grad_norm": 0.34499385952949524,
|
| 1620 |
+
"learning_rate": 0.00013033548387096774,
|
| 1621 |
+
"loss": 0.1627,
|
| 1622 |
+
"step": 5400
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 17.419354838709676,
|
| 1626 |
+
"eval_loss": 0.1536460816860199,
|
| 1627 |
+
"eval_runtime": 89.4248,
|
| 1628 |
+
"eval_samples_per_second": 212.514,
|
| 1629 |
+
"eval_steps_per_second": 4.428,
|
| 1630 |
+
"step": 5400
|
| 1631 |
+
},
|
| 1632 |
+
{
|
| 1633 |
+
"epoch": 17.580645161290324,
|
| 1634 |
+
"grad_norm": 0.4599091410636902,
|
| 1635 |
+
"learning_rate": 0.00012969032258064518,
|
| 1636 |
+
"loss": 0.1611,
|
| 1637 |
+
"step": 5450
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 17.580645161290324,
|
| 1641 |
+
"eval_loss": 0.15192009508609772,
|
| 1642 |
+
"eval_runtime": 87.9147,
|
| 1643 |
+
"eval_samples_per_second": 216.164,
|
| 1644 |
+
"eval_steps_per_second": 4.504,
|
| 1645 |
+
"step": 5450
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 17.741935483870968,
|
| 1649 |
+
"grad_norm": 0.3737597167491913,
|
| 1650 |
+
"learning_rate": 0.0001290451612903226,
|
| 1651 |
+
"loss": 0.1644,
|
| 1652 |
+
"step": 5500
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 17.741935483870968,
|
| 1656 |
+
"eval_loss": 0.15356366336345673,
|
| 1657 |
+
"eval_runtime": 87.8138,
|
| 1658 |
+
"eval_samples_per_second": 216.413,
|
| 1659 |
+
"eval_steps_per_second": 4.51,
|
| 1660 |
+
"step": 5500
|
| 1661 |
+
},
|
| 1662 |
+
{
|
| 1663 |
+
"epoch": 17.903225806451612,
|
| 1664 |
+
"grad_norm": 0.49820005893707275,
|
| 1665 |
+
"learning_rate": 0.0001284,
|
| 1666 |
+
"loss": 0.1641,
|
| 1667 |
+
"step": 5550
|
| 1668 |
+
},
|
| 1669 |
+
{
|
| 1670 |
+
"epoch": 17.903225806451612,
|
| 1671 |
+
"eval_loss": 0.15130773186683655,
|
| 1672 |
+
"eval_runtime": 89.806,
|
| 1673 |
+
"eval_samples_per_second": 211.612,
|
| 1674 |
+
"eval_steps_per_second": 4.41,
|
| 1675 |
+
"step": 5550
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 18.06451612903226,
|
| 1679 |
+
"grad_norm": 0.431963711977005,
|
| 1680 |
+
"learning_rate": 0.00012775483870967742,
|
| 1681 |
+
"loss": 0.1629,
|
| 1682 |
+
"step": 5600
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 18.06451612903226,
|
| 1686 |
+
"eval_loss": 0.15356026589870453,
|
| 1687 |
+
"eval_runtime": 90.5983,
|
| 1688 |
+
"eval_samples_per_second": 209.761,
|
| 1689 |
+
"eval_steps_per_second": 4.371,
|
| 1690 |
+
"step": 5600
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 18.225806451612904,
|
| 1694 |
+
"grad_norm": 0.41033461689949036,
|
| 1695 |
+
"learning_rate": 0.00012710967741935486,
|
| 1696 |
+
"loss": 0.1592,
|
| 1697 |
+
"step": 5650
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 18.225806451612904,
|
| 1701 |
+
"eval_loss": 0.15416103601455688,
|
| 1702 |
+
"eval_runtime": 88.2514,
|
| 1703 |
+
"eval_samples_per_second": 215.339,
|
| 1704 |
+
"eval_steps_per_second": 4.487,
|
| 1705 |
+
"step": 5650
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 18.387096774193548,
|
| 1709 |
+
"grad_norm": 0.360398530960083,
|
| 1710 |
+
"learning_rate": 0.00012646451612903227,
|
| 1711 |
+
"loss": 0.1595,
|
| 1712 |
+
"step": 5700
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 18.387096774193548,
|
| 1716 |
+
"eval_loss": 0.1532827764749527,
|
| 1717 |
+
"eval_runtime": 89.7789,
|
| 1718 |
+
"eval_samples_per_second": 211.675,
|
| 1719 |
+
"eval_steps_per_second": 4.411,
|
| 1720 |
+
"step": 5700
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 18.548387096774192,
|
| 1724 |
+
"grad_norm": 0.3443894386291504,
|
| 1725 |
+
"learning_rate": 0.00012581935483870968,
|
| 1726 |
+
"loss": 0.1627,
|
| 1727 |
+
"step": 5750
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"epoch": 18.548387096774192,
|
| 1731 |
+
"eval_loss": 0.15046313405036926,
|
| 1732 |
+
"eval_runtime": 88.4154,
|
| 1733 |
+
"eval_samples_per_second": 214.94,
|
| 1734 |
+
"eval_steps_per_second": 4.479,
|
| 1735 |
+
"step": 5750
|
| 1736 |
+
},
|
| 1737 |
+
{
|
| 1738 |
+
"epoch": 18.70967741935484,
|
| 1739 |
+
"grad_norm": 0.384339839220047,
|
| 1740 |
+
"learning_rate": 0.00012517419354838712,
|
| 1741 |
+
"loss": 0.1609,
|
| 1742 |
+
"step": 5800
|
| 1743 |
+
},
|
| 1744 |
+
{
|
| 1745 |
+
"epoch": 18.70967741935484,
|
| 1746 |
+
"eval_loss": 0.1508285254240036,
|
| 1747 |
+
"eval_runtime": 89.1892,
|
| 1748 |
+
"eval_samples_per_second": 213.075,
|
| 1749 |
+
"eval_steps_per_second": 4.44,
|
| 1750 |
+
"step": 5800
|
| 1751 |
+
},
|
| 1752 |
+
{
|
| 1753 |
+
"epoch": 18.870967741935484,
|
| 1754 |
+
"grad_norm": 0.4039391577243805,
|
| 1755 |
+
"learning_rate": 0.0001245290322580645,
|
| 1756 |
+
"loss": 0.1593,
|
| 1757 |
+
"step": 5850
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 18.870967741935484,
|
| 1761 |
+
"eval_loss": 0.15185365080833435,
|
| 1762 |
+
"eval_runtime": 87.6173,
|
| 1763 |
+
"eval_samples_per_second": 216.898,
|
| 1764 |
+
"eval_steps_per_second": 4.52,
|
| 1765 |
+
"step": 5850
|
| 1766 |
+
},
|
| 1767 |
+
{
|
| 1768 |
+
"epoch": 19.032258064516128,
|
| 1769 |
+
"grad_norm": 0.36354830861091614,
|
| 1770 |
+
"learning_rate": 0.00012388387096774195,
|
| 1771 |
+
"loss": 0.1609,
|
| 1772 |
+
"step": 5900
|
| 1773 |
+
},
|
| 1774 |
+
{
|
| 1775 |
+
"epoch": 19.032258064516128,
|
| 1776 |
+
"eval_loss": 0.15022191405296326,
|
| 1777 |
+
"eval_runtime": 88.0513,
|
| 1778 |
+
"eval_samples_per_second": 215.829,
|
| 1779 |
+
"eval_steps_per_second": 4.497,
|
| 1780 |
+
"step": 5900
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 19.193548387096776,
|
| 1784 |
+
"grad_norm": 0.41100090742111206,
|
| 1785 |
+
"learning_rate": 0.00012323870967741936,
|
| 1786 |
+
"loss": 0.1573,
|
| 1787 |
+
"step": 5950
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 19.193548387096776,
|
| 1791 |
+
"eval_loss": 0.14913968741893768,
|
| 1792 |
+
"eval_runtime": 89.4022,
|
| 1793 |
+
"eval_samples_per_second": 212.568,
|
| 1794 |
+
"eval_steps_per_second": 4.429,
|
| 1795 |
+
"step": 5950
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 19.35483870967742,
|
| 1799 |
+
"grad_norm": 0.3832205832004547,
|
| 1800 |
+
"learning_rate": 0.00012259354838709677,
|
| 1801 |
+
"loss": 0.1542,
|
| 1802 |
+
"step": 6000
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 19.35483870967742,
|
| 1806 |
+
"eval_loss": 0.14829514920711517,
|
| 1807 |
+
"eval_runtime": 86.5784,
|
| 1808 |
+
"eval_samples_per_second": 219.5,
|
| 1809 |
+
"eval_steps_per_second": 4.574,
|
| 1810 |
+
"step": 6000
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 19.516129032258064,
|
| 1814 |
+
"grad_norm": 0.3583919405937195,
|
| 1815 |
+
"learning_rate": 0.00012194838709677421,
|
| 1816 |
+
"loss": 0.1562,
|
| 1817 |
+
"step": 6050
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 19.516129032258064,
|
| 1821 |
+
"eval_loss": 0.14888739585876465,
|
| 1822 |
+
"eval_runtime": 84.8338,
|
| 1823 |
+
"eval_samples_per_second": 224.014,
|
| 1824 |
+
"eval_steps_per_second": 4.668,
|
| 1825 |
+
"step": 6050
|
| 1826 |
+
},
|
| 1827 |
+
{
|
| 1828 |
+
"epoch": 19.677419354838708,
|
| 1829 |
+
"grad_norm": 0.3783506751060486,
|
| 1830 |
+
"learning_rate": 0.00012130322580645161,
|
| 1831 |
+
"loss": 0.1551,
|
| 1832 |
+
"step": 6100
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 19.677419354838708,
|
| 1836 |
+
"eval_loss": 0.1506240963935852,
|
| 1837 |
+
"eval_runtime": 87.3622,
|
| 1838 |
+
"eval_samples_per_second": 217.531,
|
| 1839 |
+
"eval_steps_per_second": 4.533,
|
| 1840 |
+
"step": 6100
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 19.838709677419356,
|
| 1844 |
+
"grad_norm": 0.39638015627861023,
|
| 1845 |
+
"learning_rate": 0.00012065806451612905,
|
| 1846 |
+
"loss": 0.1544,
|
| 1847 |
+
"step": 6150
|
| 1848 |
+
},
|
| 1849 |
+
{
|
| 1850 |
+
"epoch": 19.838709677419356,
|
| 1851 |
+
"eval_loss": 0.1477993279695511,
|
| 1852 |
+
"eval_runtime": 90.2688,
|
| 1853 |
+
"eval_samples_per_second": 210.527,
|
| 1854 |
+
"eval_steps_per_second": 4.387,
|
| 1855 |
+
"step": 6150
|
| 1856 |
+
},
|
| 1857 |
+
{
|
| 1858 |
+
"epoch": 20.0,
|
| 1859 |
+
"grad_norm": 0.5352652072906494,
|
| 1860 |
+
"learning_rate": 0.00012001290322580645,
|
| 1861 |
+
"loss": 0.1558,
|
| 1862 |
+
"step": 6200
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"epoch": 20.0,
|
| 1866 |
+
"eval_loss": 0.1461372971534729,
|
| 1867 |
+
"eval_runtime": 88.0458,
|
| 1868 |
+
"eval_samples_per_second": 215.842,
|
| 1869 |
+
"eval_steps_per_second": 4.498,
|
| 1870 |
+
"step": 6200
|
| 1871 |
+
},
|
| 1872 |
+
{
|
| 1873 |
+
"epoch": 20.161290322580644,
|
| 1874 |
+
"grad_norm": 0.32452672719955444,
|
| 1875 |
+
"learning_rate": 0.00011936774193548387,
|
| 1876 |
+
"loss": 0.15,
|
| 1877 |
+
"step": 6250
|
| 1878 |
+
},
|
| 1879 |
+
{
|
| 1880 |
+
"epoch": 20.161290322580644,
|
| 1881 |
+
"eval_loss": 0.14649870991706848,
|
| 1882 |
+
"eval_runtime": 86.0304,
|
| 1883 |
+
"eval_samples_per_second": 220.899,
|
| 1884 |
+
"eval_steps_per_second": 4.603,
|
| 1885 |
+
"step": 6250
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 20.322580645161292,
|
| 1889 |
+
"grad_norm": 0.3544420599937439,
|
| 1890 |
+
"learning_rate": 0.00011872258064516129,
|
| 1891 |
+
"loss": 0.1451,
|
| 1892 |
+
"step": 6300
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 20.322580645161292,
|
| 1896 |
+
"eval_loss": 0.14727556705474854,
|
| 1897 |
+
"eval_runtime": 89.6974,
|
| 1898 |
+
"eval_samples_per_second": 211.868,
|
| 1899 |
+
"eval_steps_per_second": 4.415,
|
| 1900 |
+
"step": 6300
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 20.483870967741936,
|
| 1904 |
+
"grad_norm": 0.3795066773891449,
|
| 1905 |
+
"learning_rate": 0.00011807741935483871,
|
| 1906 |
+
"loss": 0.1544,
|
| 1907 |
+
"step": 6350
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 20.483870967741936,
|
| 1911 |
+
"eval_loss": 0.1442754566669464,
|
| 1912 |
+
"eval_runtime": 89.414,
|
| 1913 |
+
"eval_samples_per_second": 212.54,
|
| 1914 |
+
"eval_steps_per_second": 4.429,
|
| 1915 |
+
"step": 6350
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 20.64516129032258,
|
| 1919 |
+
"grad_norm": 0.4115369915962219,
|
| 1920 |
+
"learning_rate": 0.00011743225806451614,
|
| 1921 |
+
"loss": 0.1507,
|
| 1922 |
+
"step": 6400
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 20.64516129032258,
|
| 1926 |
+
"eval_loss": 0.14425314962863922,
|
| 1927 |
+
"eval_runtime": 88.6568,
|
| 1928 |
+
"eval_samples_per_second": 214.355,
|
| 1929 |
+
"eval_steps_per_second": 4.467,
|
| 1930 |
+
"step": 6400
|
| 1931 |
+
},
|
| 1932 |
+
{
|
| 1933 |
+
"epoch": 20.806451612903224,
|
| 1934 |
+
"grad_norm": 0.38675764203071594,
|
| 1935 |
+
"learning_rate": 0.00011678709677419355,
|
| 1936 |
+
"loss": 0.1497,
|
| 1937 |
+
"step": 6450
|
| 1938 |
+
},
|
| 1939 |
+
{
|
| 1940 |
+
"epoch": 20.806451612903224,
|
| 1941 |
+
"eval_loss": 0.1459958851337433,
|
| 1942 |
+
"eval_runtime": 87.6832,
|
| 1943 |
+
"eval_samples_per_second": 216.735,
|
| 1944 |
+
"eval_steps_per_second": 4.516,
|
| 1945 |
+
"step": 6450
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 20.967741935483872,
|
| 1949 |
+
"grad_norm": 0.37480926513671875,
|
| 1950 |
+
"learning_rate": 0.00011614193548387098,
|
| 1951 |
+
"loss": 0.1509,
|
| 1952 |
+
"step": 6500
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 20.967741935483872,
|
| 1956 |
+
"eval_loss": 0.14522159099578857,
|
| 1957 |
+
"eval_runtime": 86.8578,
|
| 1958 |
+
"eval_samples_per_second": 218.794,
|
| 1959 |
+
"eval_steps_per_second": 4.559,
|
| 1960 |
+
"step": 6500
|
| 1961 |
+
},
|
| 1962 |
+
{
|
| 1963 |
+
"epoch": 21.129032258064516,
|
| 1964 |
+
"grad_norm": 0.3740350306034088,
|
| 1965 |
+
"learning_rate": 0.00011549677419354839,
|
| 1966 |
+
"loss": 0.1504,
|
| 1967 |
+
"step": 6550
|
| 1968 |
+
},
|
| 1969 |
+
{
|
| 1970 |
+
"epoch": 21.129032258064516,
|
| 1971 |
+
"eval_loss": 0.14784836769104004,
|
| 1972 |
+
"eval_runtime": 85.4911,
|
| 1973 |
+
"eval_samples_per_second": 222.292,
|
| 1974 |
+
"eval_steps_per_second": 4.632,
|
| 1975 |
+
"step": 6550
|
| 1976 |
+
},
|
| 1977 |
+
{
|
| 1978 |
+
"epoch": 21.29032258064516,
|
| 1979 |
+
"grad_norm": 0.4533497095108032,
|
| 1980 |
+
"learning_rate": 0.00011485161290322581,
|
| 1981 |
+
"loss": 0.1517,
|
| 1982 |
+
"step": 6600
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 21.29032258064516,
|
| 1986 |
+
"eval_loss": 0.14581461250782013,
|
| 1987 |
+
"eval_runtime": 86.0863,
|
| 1988 |
+
"eval_samples_per_second": 220.755,
|
| 1989 |
+
"eval_steps_per_second": 4.6,
|
| 1990 |
+
"step": 6600
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 21.451612903225808,
|
| 1994 |
+
"grad_norm": 0.3758571743965149,
|
| 1995 |
+
"learning_rate": 0.00011420645161290323,
|
| 1996 |
+
"loss": 0.1452,
|
| 1997 |
+
"step": 6650
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 21.451612903225808,
|
| 2001 |
+
"eval_loss": 0.1412593424320221,
|
| 2002 |
+
"eval_runtime": 86.2595,
|
| 2003 |
+
"eval_samples_per_second": 220.312,
|
| 2004 |
+
"eval_steps_per_second": 4.591,
|
| 2005 |
+
"step": 6650
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 21.612903225806452,
|
| 2009 |
+
"grad_norm": 0.3700609803199768,
|
| 2010 |
+
"learning_rate": 0.00011356129032258065,
|
| 2011 |
+
"loss": 0.1461,
|
| 2012 |
+
"step": 6700
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 21.612903225806452,
|
| 2016 |
+
"eval_loss": 0.1432737410068512,
|
| 2017 |
+
"eval_runtime": 86.1269,
|
| 2018 |
+
"eval_samples_per_second": 220.651,
|
| 2019 |
+
"eval_steps_per_second": 4.598,
|
| 2020 |
+
"step": 6700
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 21.774193548387096,
|
| 2024 |
+
"grad_norm": 0.31164905428886414,
|
| 2025 |
+
"learning_rate": 0.00011291612903225808,
|
| 2026 |
+
"loss": 0.1463,
|
| 2027 |
+
"step": 6750
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 21.774193548387096,
|
| 2031 |
+
"eval_loss": 0.14101718366146088,
|
| 2032 |
+
"eval_runtime": 88.7161,
|
| 2033 |
+
"eval_samples_per_second": 214.211,
|
| 2034 |
+
"eval_steps_per_second": 4.464,
|
| 2035 |
+
"step": 6750
|
| 2036 |
+
},
|
| 2037 |
+
{
|
| 2038 |
+
"epoch": 21.93548387096774,
|
| 2039 |
+
"grad_norm": 0.3831172287464142,
|
| 2040 |
+
"learning_rate": 0.00011227096774193549,
|
| 2041 |
+
"loss": 0.1509,
|
| 2042 |
+
"step": 6800
|
| 2043 |
+
},
|
| 2044 |
+
{
|
| 2045 |
+
"epoch": 21.93548387096774,
|
| 2046 |
+
"eval_loss": 0.14286787807941437,
|
| 2047 |
+
"eval_runtime": 88.4285,
|
| 2048 |
+
"eval_samples_per_second": 214.908,
|
| 2049 |
+
"eval_steps_per_second": 4.478,
|
| 2050 |
+
"step": 6800
|
| 2051 |
+
},
|
| 2052 |
+
{
|
| 2053 |
+
"epoch": 22.096774193548388,
|
| 2054 |
+
"grad_norm": 0.3675175905227661,
|
| 2055 |
+
"learning_rate": 0.00011162580645161292,
|
| 2056 |
+
"loss": 0.1478,
|
| 2057 |
+
"step": 6850
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 22.096774193548388,
|
| 2061 |
+
"eval_loss": 0.14311262965202332,
|
| 2062 |
+
"eval_runtime": 86.1923,
|
| 2063 |
+
"eval_samples_per_second": 220.484,
|
| 2064 |
+
"eval_steps_per_second": 4.594,
|
| 2065 |
+
"step": 6850
|
| 2066 |
+
},
|
| 2067 |
+
{
|
| 2068 |
+
"epoch": 22.258064516129032,
|
| 2069 |
+
"grad_norm": 0.4077725410461426,
|
| 2070 |
+
"learning_rate": 0.00011098064516129032,
|
| 2071 |
+
"loss": 0.1455,
|
| 2072 |
+
"step": 6900
|
| 2073 |
+
},
|
| 2074 |
+
{
|
| 2075 |
+
"epoch": 22.258064516129032,
|
| 2076 |
+
"eval_loss": 0.1406867653131485,
|
| 2077 |
+
"eval_runtime": 88.1595,
|
| 2078 |
+
"eval_samples_per_second": 215.564,
|
| 2079 |
+
"eval_steps_per_second": 4.492,
|
| 2080 |
+
"step": 6900
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 22.419354838709676,
|
| 2084 |
+
"grad_norm": 0.37918218970298767,
|
| 2085 |
+
"learning_rate": 0.00011033548387096775,
|
| 2086 |
+
"loss": 0.1438,
|
| 2087 |
+
"step": 6950
|
| 2088 |
+
},
|
| 2089 |
+
{
|
| 2090 |
+
"epoch": 22.419354838709676,
|
| 2091 |
+
"eval_loss": 0.14213036000728607,
|
| 2092 |
+
"eval_runtime": 88.6161,
|
| 2093 |
+
"eval_samples_per_second": 214.453,
|
| 2094 |
+
"eval_steps_per_second": 4.469,
|
| 2095 |
+
"step": 6950
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 22.580645161290324,
|
| 2099 |
+
"grad_norm": 0.4112975597381592,
|
| 2100 |
+
"learning_rate": 0.00010969032258064518,
|
| 2101 |
+
"loss": 0.1471,
|
| 2102 |
+
"step": 7000
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 22.580645161290324,
|
| 2106 |
+
"eval_loss": 0.14302890002727509,
|
| 2107 |
+
"eval_runtime": 87.2404,
|
| 2108 |
+
"eval_samples_per_second": 217.835,
|
| 2109 |
+
"eval_steps_per_second": 4.539,
|
| 2110 |
+
"step": 7000
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 22.741935483870968,
|
| 2114 |
+
"grad_norm": 0.3555707335472107,
|
| 2115 |
+
"learning_rate": 0.00010904516129032258,
|
| 2116 |
+
"loss": 0.1435,
|
| 2117 |
+
"step": 7050
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 22.741935483870968,
|
| 2121 |
+
"eval_loss": 0.14145122468471527,
|
| 2122 |
+
"eval_runtime": 87.7191,
|
| 2123 |
+
"eval_samples_per_second": 216.646,
|
| 2124 |
+
"eval_steps_per_second": 4.514,
|
| 2125 |
+
"step": 7050
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"epoch": 22.903225806451612,
|
| 2129 |
+
"grad_norm": 0.33775362372398376,
|
| 2130 |
+
"learning_rate": 0.00010840000000000002,
|
| 2131 |
+
"loss": 0.1459,
|
| 2132 |
+
"step": 7100
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"epoch": 22.903225806451612,
|
| 2136 |
+
"eval_loss": 0.14196287095546722,
|
| 2137 |
+
"eval_runtime": 85.2736,
|
| 2138 |
+
"eval_samples_per_second": 222.859,
|
| 2139 |
+
"eval_steps_per_second": 4.644,
|
| 2140 |
+
"step": 7100
|
| 2141 |
+
},
|
| 2142 |
+
{
|
| 2143 |
+
"epoch": 23.06451612903226,
|
| 2144 |
+
"grad_norm": 0.40644168853759766,
|
| 2145 |
+
"learning_rate": 0.00010775483870967742,
|
| 2146 |
+
"loss": 0.1403,
|
| 2147 |
+
"step": 7150
|
| 2148 |
+
},
|
| 2149 |
+
{
|
| 2150 |
+
"epoch": 23.06451612903226,
|
| 2151 |
+
"eval_loss": 0.13985274732112885,
|
| 2152 |
+
"eval_runtime": 88.2679,
|
| 2153 |
+
"eval_samples_per_second": 215.299,
|
| 2154 |
+
"eval_steps_per_second": 4.486,
|
| 2155 |
+
"step": 7150
|
| 2156 |
+
},
|
| 2157 |
+
{
|
| 2158 |
+
"epoch": 23.225806451612904,
|
| 2159 |
+
"grad_norm": 0.30164963006973267,
|
| 2160 |
+
"learning_rate": 0.00010710967741935484,
|
| 2161 |
+
"loss": 0.1438,
|
| 2162 |
+
"step": 7200
|
| 2163 |
+
},
|
| 2164 |
+
{
|
| 2165 |
+
"epoch": 23.225806451612904,
|
| 2166 |
+
"eval_loss": 0.13959668576717377,
|
| 2167 |
+
"eval_runtime": 85.8815,
|
| 2168 |
+
"eval_samples_per_second": 221.282,
|
| 2169 |
+
"eval_steps_per_second": 4.611,
|
| 2170 |
+
"step": 7200
|
| 2171 |
+
},
|
| 2172 |
+
{
|
| 2173 |
+
"epoch": 23.387096774193548,
|
| 2174 |
+
"grad_norm": 0.41760918498039246,
|
| 2175 |
+
"learning_rate": 0.00010646451612903226,
|
| 2176 |
+
"loss": 0.1455,
|
| 2177 |
+
"step": 7250
|
| 2178 |
+
},
|
| 2179 |
+
{
|
| 2180 |
+
"epoch": 23.387096774193548,
|
| 2181 |
+
"eval_loss": 0.1405312865972519,
|
| 2182 |
+
"eval_runtime": 87.4804,
|
| 2183 |
+
"eval_samples_per_second": 217.237,
|
| 2184 |
+
"eval_steps_per_second": 4.527,
|
| 2185 |
+
"step": 7250
|
| 2186 |
+
},
|
| 2187 |
+
{
|
| 2188 |
+
"epoch": 23.548387096774192,
|
| 2189 |
+
"grad_norm": 0.31449875235557556,
|
| 2190 |
+
"learning_rate": 0.00010581935483870968,
|
| 2191 |
+
"loss": 0.1416,
|
| 2192 |
+
"step": 7300
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"epoch": 23.548387096774192,
|
| 2196 |
+
"eval_loss": 0.1397952139377594,
|
| 2197 |
+
"eval_runtime": 84.3693,
|
| 2198 |
+
"eval_samples_per_second": 225.248,
|
| 2199 |
+
"eval_steps_per_second": 4.694,
|
| 2200 |
+
"step": 7300
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 23.70967741935484,
|
| 2204 |
+
"grad_norm": 0.34104588627815247,
|
| 2205 |
+
"learning_rate": 0.00010517419354838711,
|
| 2206 |
+
"loss": 0.143,
|
| 2207 |
+
"step": 7350
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 23.70967741935484,
|
| 2211 |
+
"eval_loss": 0.13995403051376343,
|
| 2212 |
+
"eval_runtime": 88.9533,
|
| 2213 |
+
"eval_samples_per_second": 213.64,
|
| 2214 |
+
"eval_steps_per_second": 4.452,
|
| 2215 |
+
"step": 7350
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 23.870967741935484,
|
| 2219 |
+
"grad_norm": 0.43316343426704407,
|
| 2220 |
+
"learning_rate": 0.00010452903225806452,
|
| 2221 |
+
"loss": 0.1432,
|
| 2222 |
+
"step": 7400
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 23.870967741935484,
|
| 2226 |
+
"eval_loss": 0.13980048894882202,
|
| 2227 |
+
"eval_runtime": 87.8388,
|
| 2228 |
+
"eval_samples_per_second": 216.351,
|
| 2229 |
+
"eval_steps_per_second": 4.508,
|
| 2230 |
+
"step": 7400
|
| 2231 |
+
},
|
| 2232 |
+
{
|
| 2233 |
+
"epoch": 24.032258064516128,
|
| 2234 |
+
"grad_norm": 0.35213446617126465,
|
| 2235 |
+
"learning_rate": 0.00010388387096774195,
|
| 2236 |
+
"loss": 0.1463,
|
| 2237 |
+
"step": 7450
|
| 2238 |
+
},
|
| 2239 |
+
{
|
| 2240 |
+
"epoch": 24.032258064516128,
|
| 2241 |
+
"eval_loss": 0.1396203339099884,
|
| 2242 |
+
"eval_runtime": 86.1081,
|
| 2243 |
+
"eval_samples_per_second": 220.699,
|
| 2244 |
+
"eval_steps_per_second": 4.599,
|
| 2245 |
+
"step": 7450
|
| 2246 |
+
},
|
| 2247 |
+
{
|
| 2248 |
+
"epoch": 24.193548387096776,
|
| 2249 |
+
"grad_norm": 0.2780129313468933,
|
| 2250 |
+
"learning_rate": 0.00010323870967741936,
|
| 2251 |
+
"loss": 0.1396,
|
| 2252 |
+
"step": 7500
|
| 2253 |
+
},
|
| 2254 |
+
{
|
| 2255 |
+
"epoch": 24.193548387096776,
|
| 2256 |
+
"eval_loss": 0.13865940272808075,
|
| 2257 |
+
"eval_runtime": 84.8365,
|
| 2258 |
+
"eval_samples_per_second": 224.007,
|
| 2259 |
+
"eval_steps_per_second": 4.668,
|
| 2260 |
+
"step": 7500
|
| 2261 |
+
},
|
| 2262 |
+
{
|
| 2263 |
+
"epoch": 24.35483870967742,
|
| 2264 |
+
"grad_norm": 0.34334343671798706,
|
| 2265 |
+
"learning_rate": 0.00010259354838709679,
|
| 2266 |
+
"loss": 0.1395,
|
| 2267 |
+
"step": 7550
|
| 2268 |
+
},
|
| 2269 |
+
{
|
| 2270 |
+
"epoch": 24.35483870967742,
|
| 2271 |
+
"eval_loss": 0.1386643797159195,
|
| 2272 |
+
"eval_runtime": 85.7437,
|
| 2273 |
+
"eval_samples_per_second": 221.637,
|
| 2274 |
+
"eval_steps_per_second": 4.618,
|
| 2275 |
+
"step": 7550
|
| 2276 |
+
},
|
| 2277 |
+
{
|
| 2278 |
+
"epoch": 24.516129032258064,
|
| 2279 |
+
"grad_norm": 0.3119650185108185,
|
| 2280 |
+
"learning_rate": 0.00010194838709677418,
|
| 2281 |
+
"loss": 0.1381,
|
| 2282 |
+
"step": 7600
|
| 2283 |
+
},
|
| 2284 |
+
{
|
| 2285 |
+
"epoch": 24.516129032258064,
|
| 2286 |
+
"eval_loss": 0.1378747671842575,
|
| 2287 |
+
"eval_runtime": 94.9067,
|
| 2288 |
+
"eval_samples_per_second": 200.239,
|
| 2289 |
+
"eval_steps_per_second": 4.173,
|
| 2290 |
+
"step": 7600
|
| 2291 |
+
},
|
| 2292 |
+
{
|
| 2293 |
+
"epoch": 24.677419354838708,
|
| 2294 |
+
"grad_norm": 0.36497557163238525,
|
| 2295 |
+
"learning_rate": 0.00010130322580645162,
|
| 2296 |
+
"loss": 0.1429,
|
| 2297 |
+
"step": 7650
|
| 2298 |
+
},
|
| 2299 |
+
{
|
| 2300 |
+
"epoch": 24.677419354838708,
|
| 2301 |
+
"eval_loss": 0.1373891532421112,
|
| 2302 |
+
"eval_runtime": 86.0356,
|
| 2303 |
+
"eval_samples_per_second": 220.885,
|
| 2304 |
+
"eval_steps_per_second": 4.603,
|
| 2305 |
+
"step": 7650
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 24.838709677419356,
|
| 2309 |
+
"grad_norm": 0.3456083834171295,
|
| 2310 |
+
"learning_rate": 0.00010065806451612905,
|
| 2311 |
+
"loss": 0.1349,
|
| 2312 |
+
"step": 7700
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 24.838709677419356,
|
| 2316 |
+
"eval_loss": 0.1388697326183319,
|
| 2317 |
+
"eval_runtime": 85.5136,
|
| 2318 |
+
"eval_samples_per_second": 222.234,
|
| 2319 |
+
"eval_steps_per_second": 4.631,
|
| 2320 |
+
"step": 7700
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 25.0,
|
| 2324 |
+
"grad_norm": 0.4443909227848053,
|
| 2325 |
+
"learning_rate": 0.00010001290322580645,
|
| 2326 |
+
"loss": 0.1403,
|
| 2327 |
+
"step": 7750
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 25.0,
|
| 2331 |
+
"eval_loss": 0.1377411037683487,
|
| 2332 |
+
"eval_runtime": 91.0633,
|
| 2333 |
+
"eval_samples_per_second": 208.69,
|
| 2334 |
+
"eval_steps_per_second": 4.349,
|
| 2335 |
+
"step": 7750
|
| 2336 |
+
},
|
| 2337 |
+
{
|
| 2338 |
+
"epoch": 25.161290322580644,
|
| 2339 |
+
"grad_norm": 0.34777510166168213,
|
| 2340 |
+
"learning_rate": 9.936774193548387e-05,
|
| 2341 |
+
"loss": 0.1383,
|
| 2342 |
+
"step": 7800
|
| 2343 |
+
},
|
| 2344 |
+
{
|
| 2345 |
+
"epoch": 25.161290322580644,
|
| 2346 |
+
"eval_loss": 0.13685546815395355,
|
| 2347 |
+
"eval_runtime": 88.0711,
|
| 2348 |
+
"eval_samples_per_second": 215.78,
|
| 2349 |
+
"eval_steps_per_second": 4.496,
|
| 2350 |
+
"step": 7800
|
| 2351 |
+
},
|
| 2352 |
+
{
|
| 2353 |
+
"epoch": 25.322580645161292,
|
| 2354 |
+
"grad_norm": 0.35419756174087524,
|
| 2355 |
+
"learning_rate": 9.87225806451613e-05,
|
| 2356 |
+
"loss": 0.1348,
|
| 2357 |
+
"step": 7850
|
| 2358 |
+
},
|
| 2359 |
+
{
|
| 2360 |
+
"epoch": 25.322580645161292,
|
| 2361 |
+
"eval_loss": 0.13754527270793915,
|
| 2362 |
+
"eval_runtime": 85.9879,
|
| 2363 |
+
"eval_samples_per_second": 221.008,
|
| 2364 |
+
"eval_steps_per_second": 4.605,
|
| 2365 |
+
"step": 7850
|
| 2366 |
+
},
|
| 2367 |
+
{
|
| 2368 |
+
"epoch": 25.483870967741936,
|
| 2369 |
+
"grad_norm": 0.35281285643577576,
|
| 2370 |
+
"learning_rate": 9.807741935483871e-05,
|
| 2371 |
+
"loss": 0.136,
|
| 2372 |
+
"step": 7900
|
| 2373 |
+
},
|
| 2374 |
+
{
|
| 2375 |
+
"epoch": 25.483870967741936,
|
| 2376 |
+
"eval_loss": 0.13788650929927826,
|
| 2377 |
+
"eval_runtime": 86.623,
|
| 2378 |
+
"eval_samples_per_second": 219.387,
|
| 2379 |
+
"eval_steps_per_second": 4.572,
|
| 2380 |
+
"step": 7900
|
| 2381 |
+
},
|
| 2382 |
+
{
|
| 2383 |
+
"epoch": 25.64516129032258,
|
| 2384 |
+
"grad_norm": 0.26881253719329834,
|
| 2385 |
+
"learning_rate": 9.743225806451614e-05,
|
| 2386 |
+
"loss": 0.1376,
|
| 2387 |
+
"step": 7950
|
| 2388 |
+
},
|
| 2389 |
+
{
|
| 2390 |
+
"epoch": 25.64516129032258,
|
| 2391 |
+
"eval_loss": 0.13465990126132965,
|
| 2392 |
+
"eval_runtime": 87.1032,
|
| 2393 |
+
"eval_samples_per_second": 218.178,
|
| 2394 |
+
"eval_steps_per_second": 4.546,
|
| 2395 |
+
"step": 7950
|
| 2396 |
+
},
|
| 2397 |
+
{
|
| 2398 |
+
"epoch": 25.806451612903224,
|
| 2399 |
+
"grad_norm": 0.38799649477005005,
|
| 2400 |
+
"learning_rate": 9.678709677419355e-05,
|
| 2401 |
+
"loss": 0.1365,
|
| 2402 |
+
"step": 8000
|
| 2403 |
+
},
|
| 2404 |
+
{
|
| 2405 |
+
"epoch": 25.806451612903224,
|
| 2406 |
+
"eval_loss": 0.13522003591060638,
|
| 2407 |
+
"eval_runtime": 95.2991,
|
| 2408 |
+
"eval_samples_per_second": 199.414,
|
| 2409 |
+
"eval_steps_per_second": 4.155,
|
| 2410 |
+
"step": 8000
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 25.967741935483872,
|
| 2414 |
+
"grad_norm": 0.37531042098999023,
|
| 2415 |
+
"learning_rate": 9.614193548387098e-05,
|
| 2416 |
+
"loss": 0.1362,
|
| 2417 |
+
"step": 8050
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 25.967741935483872,
|
| 2421 |
+
"eval_loss": 0.13398829102516174,
|
| 2422 |
+
"eval_runtime": 87.8409,
|
| 2423 |
+
"eval_samples_per_second": 216.346,
|
| 2424 |
+
"eval_steps_per_second": 4.508,
|
| 2425 |
+
"step": 8050
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 26.129032258064516,
|
| 2429 |
+
"grad_norm": 0.3436211049556732,
|
| 2430 |
+
"learning_rate": 9.549677419354839e-05,
|
| 2431 |
+
"loss": 0.1342,
|
| 2432 |
+
"step": 8100
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 26.129032258064516,
|
| 2436 |
+
"eval_loss": 0.13594096899032593,
|
| 2437 |
+
"eval_runtime": 84.6121,
|
| 2438 |
+
"eval_samples_per_second": 224.602,
|
| 2439 |
+
"eval_steps_per_second": 4.68,
|
| 2440 |
+
"step": 8100
|
| 2441 |
+
},
|
| 2442 |
+
{
|
| 2443 |
+
"epoch": 26.29032258064516,
|
| 2444 |
+
"grad_norm": 0.38407161831855774,
|
| 2445 |
+
"learning_rate": 9.48516129032258e-05,
|
| 2446 |
+
"loss": 0.1322,
|
| 2447 |
+
"step": 8150
|
| 2448 |
+
},
|
| 2449 |
+
{
|
| 2450 |
+
"epoch": 26.29032258064516,
|
| 2451 |
+
"eval_loss": 0.13434267044067383,
|
| 2452 |
+
"eval_runtime": 86.2264,
|
| 2453 |
+
"eval_samples_per_second": 220.396,
|
| 2454 |
+
"eval_steps_per_second": 4.593,
|
| 2455 |
+
"step": 8150
|
| 2456 |
+
},
|
| 2457 |
+
{
|
| 2458 |
+
"epoch": 26.451612903225808,
|
| 2459 |
+
"grad_norm": 0.3329039514064789,
|
| 2460 |
+
"learning_rate": 9.420645161290324e-05,
|
| 2461 |
+
"loss": 0.1338,
|
| 2462 |
+
"step": 8200
|
| 2463 |
+
},
|
| 2464 |
+
{
|
| 2465 |
+
"epoch": 26.451612903225808,
|
| 2466 |
+
"eval_loss": 0.1350966840982437,
|
| 2467 |
+
"eval_runtime": 87.8388,
|
| 2468 |
+
"eval_samples_per_second": 216.351,
|
| 2469 |
+
"eval_steps_per_second": 4.508,
|
| 2470 |
+
"step": 8200
|
| 2471 |
+
},
|
| 2472 |
+
{
|
| 2473 |
+
"epoch": 26.612903225806452,
|
| 2474 |
+
"grad_norm": 0.41340529918670654,
|
| 2475 |
+
"learning_rate": 9.356129032258065e-05,
|
| 2476 |
+
"loss": 0.1362,
|
| 2477 |
+
"step": 8250
|
| 2478 |
+
},
|
| 2479 |
+
{
|
| 2480 |
+
"epoch": 26.612903225806452,
|
| 2481 |
+
"eval_loss": 0.13463687896728516,
|
| 2482 |
+
"eval_runtime": 86.5564,
|
| 2483 |
+
"eval_samples_per_second": 219.556,
|
| 2484 |
+
"eval_steps_per_second": 4.575,
|
| 2485 |
+
"step": 8250
|
| 2486 |
+
},
|
| 2487 |
+
{
|
| 2488 |
+
"epoch": 26.774193548387096,
|
| 2489 |
+
"grad_norm": 0.33343157172203064,
|
| 2490 |
+
"learning_rate": 9.291612903225807e-05,
|
| 2491 |
+
"loss": 0.1331,
|
| 2492 |
+
"step": 8300
|
| 2493 |
+
},
|
| 2494 |
+
{
|
| 2495 |
+
"epoch": 26.774193548387096,
|
| 2496 |
+
"eval_loss": 0.1351860612630844,
|
| 2497 |
+
"eval_runtime": 88.4351,
|
| 2498 |
+
"eval_samples_per_second": 214.892,
|
| 2499 |
+
"eval_steps_per_second": 4.478,
|
| 2500 |
+
"step": 8300
|
| 2501 |
+
},
|
| 2502 |
+
{
|
| 2503 |
+
"epoch": 26.93548387096774,
|
| 2504 |
+
"grad_norm": 0.3529933989048004,
|
| 2505 |
+
"learning_rate": 9.227096774193549e-05,
|
| 2506 |
+
"loss": 0.1306,
|
| 2507 |
+
"step": 8350
|
| 2508 |
+
},
|
| 2509 |
+
{
|
| 2510 |
+
"epoch": 26.93548387096774,
|
| 2511 |
+
"eval_loss": 0.13507899641990662,
|
| 2512 |
+
"eval_runtime": 88.8438,
|
| 2513 |
+
"eval_samples_per_second": 213.904,
|
| 2514 |
+
"eval_steps_per_second": 4.457,
|
| 2515 |
+
"step": 8350
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 27.096774193548388,
|
| 2519 |
+
"grad_norm": 0.3433696925640106,
|
| 2520 |
+
"learning_rate": 9.16258064516129e-05,
|
| 2521 |
+
"loss": 0.1339,
|
| 2522 |
+
"step": 8400
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 27.096774193548388,
|
| 2526 |
+
"eval_loss": 0.13309802114963531,
|
| 2527 |
+
"eval_runtime": 89.0663,
|
| 2528 |
+
"eval_samples_per_second": 213.369,
|
| 2529 |
+
"eval_steps_per_second": 4.446,
|
| 2530 |
+
"step": 8400
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 27.258064516129032,
|
| 2534 |
+
"grad_norm": 0.3371010720729828,
|
| 2535 |
+
"learning_rate": 9.098064516129032e-05,
|
| 2536 |
+
"loss": 0.1315,
|
| 2537 |
+
"step": 8450
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 27.258064516129032,
|
| 2541 |
+
"eval_loss": 0.1346307247877121,
|
| 2542 |
+
"eval_runtime": 88.7761,
|
| 2543 |
+
"eval_samples_per_second": 214.067,
|
| 2544 |
+
"eval_steps_per_second": 4.461,
|
| 2545 |
+
"step": 8450
|
| 2546 |
+
},
|
| 2547 |
+
{
|
| 2548 |
+
"epoch": 27.419354838709676,
|
| 2549 |
+
"grad_norm": 0.36965006589889526,
|
| 2550 |
+
"learning_rate": 9.033548387096774e-05,
|
| 2551 |
+
"loss": 0.1349,
|
| 2552 |
+
"step": 8500
|
| 2553 |
+
},
|
| 2554 |
+
{
|
| 2555 |
+
"epoch": 27.419354838709676,
|
| 2556 |
+
"eval_loss": 0.13345304131507874,
|
| 2557 |
+
"eval_runtime": 89.2673,
|
| 2558 |
+
"eval_samples_per_second": 212.889,
|
| 2559 |
+
"eval_steps_per_second": 4.436,
|
| 2560 |
+
"step": 8500
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 27.580645161290324,
|
| 2564 |
+
"grad_norm": 0.3361060917377472,
|
| 2565 |
+
"learning_rate": 8.969032258064517e-05,
|
| 2566 |
+
"loss": 0.1323,
|
| 2567 |
+
"step": 8550
|
| 2568 |
+
},
|
| 2569 |
+
{
|
| 2570 |
+
"epoch": 27.580645161290324,
|
| 2571 |
+
"eval_loss": 0.1327604055404663,
|
| 2572 |
+
"eval_runtime": 87.1092,
|
| 2573 |
+
"eval_samples_per_second": 218.163,
|
| 2574 |
+
"eval_steps_per_second": 4.546,
|
| 2575 |
+
"step": 8550
|
| 2576 |
+
},
|
| 2577 |
+
{
|
| 2578 |
+
"epoch": 27.741935483870968,
|
| 2579 |
+
"grad_norm": 0.2936784029006958,
|
| 2580 |
+
"learning_rate": 8.904516129032258e-05,
|
| 2581 |
+
"loss": 0.132,
|
| 2582 |
+
"step": 8600
|
| 2583 |
+
},
|
| 2584 |
+
{
|
| 2585 |
+
"epoch": 27.741935483870968,
|
| 2586 |
+
"eval_loss": 0.1343098133802414,
|
| 2587 |
+
"eval_runtime": 89.5037,
|
| 2588 |
+
"eval_samples_per_second": 212.326,
|
| 2589 |
+
"eval_steps_per_second": 4.424,
|
| 2590 |
+
"step": 8600
|
| 2591 |
+
},
|
| 2592 |
+
{
|
| 2593 |
+
"epoch": 27.903225806451612,
|
| 2594 |
+
"grad_norm": 0.332289457321167,
|
| 2595 |
+
"learning_rate": 8.840000000000001e-05,
|
| 2596 |
+
"loss": 0.1303,
|
| 2597 |
+
"step": 8650
|
| 2598 |
+
},
|
| 2599 |
+
{
|
| 2600 |
+
"epoch": 27.903225806451612,
|
| 2601 |
+
"eval_loss": 0.1309853345155716,
|
| 2602 |
+
"eval_runtime": 88.6535,
|
| 2603 |
+
"eval_samples_per_second": 214.363,
|
| 2604 |
+
"eval_steps_per_second": 4.467,
|
| 2605 |
+
"step": 8650
|
| 2606 |
+
},
|
| 2607 |
+
{
|
| 2608 |
+
"epoch": 28.06451612903226,
|
| 2609 |
+
"grad_norm": 0.3243560791015625,
|
| 2610 |
+
"learning_rate": 8.775483870967742e-05,
|
| 2611 |
+
"loss": 0.1295,
|
| 2612 |
+
"step": 8700
|
| 2613 |
+
},
|
| 2614 |
+
{
|
| 2615 |
+
"epoch": 28.06451612903226,
|
| 2616 |
+
"eval_loss": 0.13385291397571564,
|
| 2617 |
+
"eval_runtime": 89.1997,
|
| 2618 |
+
"eval_samples_per_second": 213.05,
|
| 2619 |
+
"eval_steps_per_second": 4.439,
|
| 2620 |
+
"step": 8700
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 28.225806451612904,
|
| 2624 |
+
"grad_norm": 0.2707726061344147,
|
| 2625 |
+
"learning_rate": 8.710967741935485e-05,
|
| 2626 |
+
"loss": 0.1296,
|
| 2627 |
+
"step": 8750
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 28.225806451612904,
|
| 2631 |
+
"eval_loss": 0.131495863199234,
|
| 2632 |
+
"eval_runtime": 87.7498,
|
| 2633 |
+
"eval_samples_per_second": 216.57,
|
| 2634 |
+
"eval_steps_per_second": 4.513,
|
| 2635 |
+
"step": 8750
|
| 2636 |
+
},
|
| 2637 |
+
{
|
| 2638 |
+
"epoch": 28.387096774193548,
|
| 2639 |
+
"grad_norm": 0.3294861912727356,
|
| 2640 |
+
"learning_rate": 8.646451612903226e-05,
|
| 2641 |
+
"loss": 0.1309,
|
| 2642 |
+
"step": 8800
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 28.387096774193548,
|
| 2646 |
+
"eval_loss": 0.1315741389989853,
|
| 2647 |
+
"eval_runtime": 89.3762,
|
| 2648 |
+
"eval_samples_per_second": 212.629,
|
| 2649 |
+
"eval_steps_per_second": 4.431,
|
| 2650 |
+
"step": 8800
|
| 2651 |
+
},
|
| 2652 |
+
{
|
| 2653 |
+
"epoch": 28.548387096774192,
|
| 2654 |
+
"grad_norm": 0.3417121469974518,
|
| 2655 |
+
"learning_rate": 8.581935483870968e-05,
|
| 2656 |
+
"loss": 0.1259,
|
| 2657 |
+
"step": 8850
|
| 2658 |
+
},
|
| 2659 |
+
{
|
| 2660 |
+
"epoch": 28.548387096774192,
|
| 2661 |
+
"eval_loss": 0.13167841732501984,
|
| 2662 |
+
"eval_runtime": 89.5004,
|
| 2663 |
+
"eval_samples_per_second": 212.334,
|
| 2664 |
+
"eval_steps_per_second": 4.425,
|
| 2665 |
+
"step": 8850
|
| 2666 |
+
},
|
| 2667 |
+
{
|
| 2668 |
+
"epoch": 28.70967741935484,
|
| 2669 |
+
"grad_norm": 0.4122408628463745,
|
| 2670 |
+
"learning_rate": 8.517419354838711e-05,
|
| 2671 |
+
"loss": 0.1309,
|
| 2672 |
+
"step": 8900
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"epoch": 28.70967741935484,
|
| 2676 |
+
"eval_loss": 0.1316230446100235,
|
| 2677 |
+
"eval_runtime": 87.7329,
|
| 2678 |
+
"eval_samples_per_second": 216.612,
|
| 2679 |
+
"eval_steps_per_second": 4.514,
|
| 2680 |
+
"step": 8900
|
| 2681 |
+
},
|
| 2682 |
+
{
|
| 2683 |
+
"epoch": 28.870967741935484,
|
| 2684 |
+
"grad_norm": 0.28204530477523804,
|
| 2685 |
+
"learning_rate": 8.452903225806452e-05,
|
| 2686 |
+
"loss": 0.1275,
|
| 2687 |
+
"step": 8950
|
| 2688 |
+
},
|
| 2689 |
+
{
|
| 2690 |
+
"epoch": 28.870967741935484,
|
| 2691 |
+
"eval_loss": 0.13107319176197052,
|
| 2692 |
+
"eval_runtime": 89.6188,
|
| 2693 |
+
"eval_samples_per_second": 212.054,
|
| 2694 |
+
"eval_steps_per_second": 4.419,
|
| 2695 |
+
"step": 8950
|
| 2696 |
+
},
|
| 2697 |
+
{
|
| 2698 |
+
"epoch": 29.032258064516128,
|
| 2699 |
+
"grad_norm": 0.35629284381866455,
|
| 2700 |
+
"learning_rate": 8.388387096774194e-05,
|
| 2701 |
+
"loss": 0.1288,
|
| 2702 |
+
"step": 9000
|
| 2703 |
+
},
|
| 2704 |
+
{
|
| 2705 |
+
"epoch": 29.032258064516128,
|
| 2706 |
+
"eval_loss": 0.13227057456970215,
|
| 2707 |
+
"eval_runtime": 88.8381,
|
| 2708 |
+
"eval_samples_per_second": 213.917,
|
| 2709 |
+
"eval_steps_per_second": 4.458,
|
| 2710 |
+
"step": 9000
|
| 2711 |
+
},
|
| 2712 |
+
{
|
| 2713 |
+
"epoch": 29.193548387096776,
|
| 2714 |
+
"grad_norm": 0.309741348028183,
|
| 2715 |
+
"learning_rate": 8.323870967741936e-05,
|
| 2716 |
+
"loss": 0.128,
|
| 2717 |
+
"step": 9050
|
| 2718 |
+
},
|
| 2719 |
+
{
|
| 2720 |
+
"epoch": 29.193548387096776,
|
| 2721 |
+
"eval_loss": 0.13167747855186462,
|
| 2722 |
+
"eval_runtime": 88.4882,
|
| 2723 |
+
"eval_samples_per_second": 214.763,
|
| 2724 |
+
"eval_steps_per_second": 4.475,
|
| 2725 |
+
"step": 9050
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 29.35483870967742,
|
| 2729 |
+
"grad_norm": 0.288798987865448,
|
| 2730 |
+
"learning_rate": 8.259354838709677e-05,
|
| 2731 |
+
"loss": 0.1267,
|
| 2732 |
+
"step": 9100
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 29.35483870967742,
|
| 2736 |
+
"eval_loss": 0.1299341917037964,
|
| 2737 |
+
"eval_runtime": 86.532,
|
| 2738 |
+
"eval_samples_per_second": 219.618,
|
| 2739 |
+
"eval_steps_per_second": 4.576,
|
| 2740 |
+
"step": 9100
|
| 2741 |
+
},
|
| 2742 |
+
{
|
| 2743 |
+
"epoch": 29.516129032258064,
|
| 2744 |
+
"grad_norm": 0.3058416545391083,
|
| 2745 |
+
"learning_rate": 8.19483870967742e-05,
|
| 2746 |
+
"loss": 0.1262,
|
| 2747 |
+
"step": 9150
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 29.516129032258064,
|
| 2751 |
+
"eval_loss": 0.1303175687789917,
|
| 2752 |
+
"eval_runtime": 87.4,
|
| 2753 |
+
"eval_samples_per_second": 217.437,
|
| 2754 |
+
"eval_steps_per_second": 4.531,
|
| 2755 |
+
"step": 9150
|
| 2756 |
+
},
|
| 2757 |
+
{
|
| 2758 |
+
"epoch": 29.677419354838708,
|
| 2759 |
+
"grad_norm": 0.357373982667923,
|
| 2760 |
+
"learning_rate": 8.130322580645163e-05,
|
| 2761 |
+
"loss": 0.1317,
|
| 2762 |
+
"step": 9200
|
| 2763 |
+
},
|
| 2764 |
+
{
|
| 2765 |
+
"epoch": 29.677419354838708,
|
| 2766 |
+
"eval_loss": 0.1297539323568344,
|
| 2767 |
+
"eval_runtime": 87.917,
|
| 2768 |
+
"eval_samples_per_second": 216.158,
|
| 2769 |
+
"eval_steps_per_second": 4.504,
|
| 2770 |
+
"step": 9200
|
| 2771 |
+
},
|
| 2772 |
+
{
|
| 2773 |
+
"epoch": 29.838709677419356,
|
| 2774 |
+
"grad_norm": 0.3070197105407715,
|
| 2775 |
+
"learning_rate": 8.065806451612904e-05,
|
| 2776 |
+
"loss": 0.1277,
|
| 2777 |
+
"step": 9250
|
| 2778 |
+
},
|
| 2779 |
+
{
|
| 2780 |
+
"epoch": 29.838709677419356,
|
| 2781 |
+
"eval_loss": 0.12903046607971191,
|
| 2782 |
+
"eval_runtime": 90.6017,
|
| 2783 |
+
"eval_samples_per_second": 209.753,
|
| 2784 |
+
"eval_steps_per_second": 4.371,
|
| 2785 |
+
"step": 9250
|
| 2786 |
+
},
|
| 2787 |
+
{
|
| 2788 |
+
"epoch": 30.0,
|
| 2789 |
+
"grad_norm": 0.36131608486175537,
|
| 2790 |
+
"learning_rate": 8.001290322580646e-05,
|
| 2791 |
+
"loss": 0.1244,
|
| 2792 |
+
"step": 9300
|
| 2793 |
+
},
|
| 2794 |
+
{
|
| 2795 |
+
"epoch": 30.0,
|
| 2796 |
+
"eval_loss": 0.1285567581653595,
|
| 2797 |
+
"eval_runtime": 88.2187,
|
| 2798 |
+
"eval_samples_per_second": 215.419,
|
| 2799 |
+
"eval_steps_per_second": 4.489,
|
| 2800 |
+
"step": 9300
|
| 2801 |
+
},
|
| 2802 |
+
{
|
| 2803 |
+
"epoch": 30.161290322580644,
|
| 2804 |
+
"grad_norm": 0.31005364656448364,
|
| 2805 |
+
"learning_rate": 7.936774193548388e-05,
|
| 2806 |
+
"loss": 0.1274,
|
| 2807 |
+
"step": 9350
|
| 2808 |
+
},
|
| 2809 |
+
{
|
| 2810 |
+
"epoch": 30.161290322580644,
|
| 2811 |
+
"eval_loss": 0.12861816585063934,
|
| 2812 |
+
"eval_runtime": 88.8974,
|
| 2813 |
+
"eval_samples_per_second": 213.775,
|
| 2814 |
+
"eval_steps_per_second": 4.455,
|
| 2815 |
+
"step": 9350
|
| 2816 |
+
},
|
| 2817 |
+
{
|
| 2818 |
+
"epoch": 30.322580645161292,
|
| 2819 |
+
"grad_norm": 0.3450087904930115,
|
| 2820 |
+
"learning_rate": 7.872258064516129e-05,
|
| 2821 |
+
"loss": 0.1315,
|
| 2822 |
+
"step": 9400
|
| 2823 |
+
},
|
| 2824 |
+
{
|
| 2825 |
+
"epoch": 30.322580645161292,
|
| 2826 |
+
"eval_loss": 0.12834839522838593,
|
| 2827 |
+
"eval_runtime": 88.9142,
|
| 2828 |
+
"eval_samples_per_second": 213.734,
|
| 2829 |
+
"eval_steps_per_second": 4.454,
|
| 2830 |
+
"step": 9400
|
| 2831 |
+
},
|
| 2832 |
+
{
|
| 2833 |
+
"epoch": 30.483870967741936,
|
| 2834 |
+
"grad_norm": 0.26987212896347046,
|
| 2835 |
+
"learning_rate": 7.807741935483871e-05,
|
| 2836 |
+
"loss": 0.1228,
|
| 2837 |
+
"step": 9450
|
| 2838 |
+
},
|
| 2839 |
+
{
|
| 2840 |
+
"epoch": 30.483870967741936,
|
| 2841 |
+
"eval_loss": 0.12817350029945374,
|
| 2842 |
+
"eval_runtime": 88.7737,
|
| 2843 |
+
"eval_samples_per_second": 214.073,
|
| 2844 |
+
"eval_steps_per_second": 4.461,
|
| 2845 |
+
"step": 9450
|
| 2846 |
+
},
|
| 2847 |
+
{
|
| 2848 |
+
"epoch": 30.64516129032258,
|
| 2849 |
+
"grad_norm": 0.3717745244503021,
|
| 2850 |
+
"learning_rate": 7.743225806451613e-05,
|
| 2851 |
+
"loss": 0.124,
|
| 2852 |
+
"step": 9500
|
| 2853 |
+
},
|
| 2854 |
+
{
|
| 2855 |
+
"epoch": 30.64516129032258,
|
| 2856 |
+
"eval_loss": 0.13065199553966522,
|
| 2857 |
+
"eval_runtime": 88.4187,
|
| 2858 |
+
"eval_samples_per_second": 214.932,
|
| 2859 |
+
"eval_steps_per_second": 4.479,
|
| 2860 |
+
"step": 9500
|
| 2861 |
+
},
|
| 2862 |
+
{
|
| 2863 |
+
"epoch": 30.806451612903224,
|
| 2864 |
+
"grad_norm": 0.29905572533607483,
|
| 2865 |
+
"learning_rate": 7.678709677419355e-05,
|
| 2866 |
+
"loss": 0.1212,
|
| 2867 |
+
"step": 9550
|
| 2868 |
+
},
|
| 2869 |
+
{
|
| 2870 |
+
"epoch": 30.806451612903224,
|
| 2871 |
+
"eval_loss": 0.1283879578113556,
|
| 2872 |
+
"eval_runtime": 87.9276,
|
| 2873 |
+
"eval_samples_per_second": 216.132,
|
| 2874 |
+
"eval_steps_per_second": 4.504,
|
| 2875 |
+
"step": 9550
|
| 2876 |
+
},
|
| 2877 |
+
{
|
| 2878 |
+
"epoch": 30.967741935483872,
|
| 2879 |
+
"grad_norm": 0.2816069722175598,
|
| 2880 |
+
"learning_rate": 7.614193548387098e-05,
|
| 2881 |
+
"loss": 0.1233,
|
| 2882 |
+
"step": 9600
|
| 2883 |
+
},
|
| 2884 |
+
{
|
| 2885 |
+
"epoch": 30.967741935483872,
|
| 2886 |
+
"eval_loss": 0.12923495471477509,
|
| 2887 |
+
"eval_runtime": 89.2797,
|
| 2888 |
+
"eval_samples_per_second": 212.859,
|
| 2889 |
+
"eval_steps_per_second": 4.436,
|
| 2890 |
+
"step": 9600
|
| 2891 |
+
},
|
| 2892 |
+
{
|
| 2893 |
+
"epoch": 31.129032258064516,
|
| 2894 |
+
"grad_norm": 0.30339810252189636,
|
| 2895 |
+
"learning_rate": 7.549677419354839e-05,
|
| 2896 |
+
"loss": 0.1226,
|
| 2897 |
+
"step": 9650
|
| 2898 |
+
},
|
| 2899 |
+
{
|
| 2900 |
+
"epoch": 31.129032258064516,
|
| 2901 |
+
"eval_loss": 0.12811945378780365,
|
| 2902 |
+
"eval_runtime": 85.4363,
|
| 2903 |
+
"eval_samples_per_second": 222.435,
|
| 2904 |
+
"eval_steps_per_second": 4.635,
|
| 2905 |
+
"step": 9650
|
| 2906 |
+
},
|
| 2907 |
+
{
|
| 2908 |
+
"epoch": 31.29032258064516,
|
| 2909 |
+
"grad_norm": 0.3012908697128296,
|
| 2910 |
+
"learning_rate": 7.48516129032258e-05,
|
| 2911 |
+
"loss": 0.1244,
|
| 2912 |
+
"step": 9700
|
| 2913 |
+
},
|
| 2914 |
+
{
|
| 2915 |
+
"epoch": 31.29032258064516,
|
| 2916 |
+
"eval_loss": 0.12850458920001984,
|
| 2917 |
+
"eval_runtime": 90.0213,
|
| 2918 |
+
"eval_samples_per_second": 211.106,
|
| 2919 |
+
"eval_steps_per_second": 4.399,
|
| 2920 |
+
"step": 9700
|
| 2921 |
+
},
|
| 2922 |
+
{
|
| 2923 |
+
"epoch": 31.451612903225808,
|
| 2924 |
+
"grad_norm": 0.36201730370521545,
|
| 2925 |
+
"learning_rate": 7.420645161290323e-05,
|
| 2926 |
+
"loss": 0.1272,
|
| 2927 |
+
"step": 9750
|
| 2928 |
+
},
|
| 2929 |
+
{
|
| 2930 |
+
"epoch": 31.451612903225808,
|
| 2931 |
+
"eval_loss": 0.12800218164920807,
|
| 2932 |
+
"eval_runtime": 95.4408,
|
| 2933 |
+
"eval_samples_per_second": 199.118,
|
| 2934 |
+
"eval_steps_per_second": 4.149,
|
| 2935 |
+
"step": 9750
|
| 2936 |
+
},
|
| 2937 |
+
{
|
| 2938 |
+
"epoch": 31.612903225806452,
|
| 2939 |
+
"grad_norm": 0.30312639474868774,
|
| 2940 |
+
"learning_rate": 7.356129032258064e-05,
|
| 2941 |
+
"loss": 0.1248,
|
| 2942 |
+
"step": 9800
|
| 2943 |
+
},
|
| 2944 |
+
{
|
| 2945 |
+
"epoch": 31.612903225806452,
|
| 2946 |
+
"eval_loss": 0.12891393899917603,
|
| 2947 |
+
"eval_runtime": 89.5417,
|
| 2948 |
+
"eval_samples_per_second": 212.236,
|
| 2949 |
+
"eval_steps_per_second": 4.423,
|
| 2950 |
+
"step": 9800
|
| 2951 |
+
},
|
| 2952 |
+
{
|
| 2953 |
+
"epoch": 31.774193548387096,
|
| 2954 |
+
"grad_norm": 0.34564414620399475,
|
| 2955 |
+
"learning_rate": 7.291612903225807e-05,
|
| 2956 |
+
"loss": 0.1245,
|
| 2957 |
+
"step": 9850
|
| 2958 |
+
},
|
| 2959 |
+
{
|
| 2960 |
+
"epoch": 31.774193548387096,
|
| 2961 |
+
"eval_loss": 0.12625598907470703,
|
| 2962 |
+
"eval_runtime": 92.1876,
|
| 2963 |
+
"eval_samples_per_second": 206.145,
|
| 2964 |
+
"eval_steps_per_second": 4.296,
|
| 2965 |
+
"step": 9850
|
| 2966 |
+
},
|
| 2967 |
+
{
|
| 2968 |
+
"epoch": 31.93548387096774,
|
| 2969 |
+
"grad_norm": 0.31116342544555664,
|
| 2970 |
+
"learning_rate": 7.22709677419355e-05,
|
| 2971 |
+
"loss": 0.1261,
|
| 2972 |
+
"step": 9900
|
| 2973 |
+
},
|
| 2974 |
+
{
|
| 2975 |
+
"epoch": 31.93548387096774,
|
| 2976 |
+
"eval_loss": 0.1266312599182129,
|
| 2977 |
+
"eval_runtime": 105.6265,
|
| 2978 |
+
"eval_samples_per_second": 179.917,
|
| 2979 |
+
"eval_steps_per_second": 3.749,
|
| 2980 |
+
"step": 9900
|
| 2981 |
+
},
|
| 2982 |
+
{
|
| 2983 |
+
"epoch": 32.096774193548384,
|
| 2984 |
+
"grad_norm": 0.2931393086910248,
|
| 2985 |
+
"learning_rate": 7.16258064516129e-05,
|
| 2986 |
+
"loss": 0.1228,
|
| 2987 |
+
"step": 9950
|
| 2988 |
+
},
|
| 2989 |
+
{
|
| 2990 |
+
"epoch": 32.096774193548384,
|
| 2991 |
+
"eval_loss": 0.12575842440128326,
|
| 2992 |
+
"eval_runtime": 89.3508,
|
| 2993 |
+
"eval_samples_per_second": 212.69,
|
| 2994 |
+
"eval_steps_per_second": 4.432,
|
| 2995 |
+
"step": 9950
|
| 2996 |
+
},
|
| 2997 |
+
{
|
| 2998 |
+
"epoch": 32.25806451612903,
|
| 2999 |
+
"grad_norm": 0.244206041097641,
|
| 3000 |
+
"learning_rate": 7.098064516129033e-05,
|
| 3001 |
+
"loss": 0.1232,
|
| 3002 |
+
"step": 10000
|
| 3003 |
+
},
|
| 3004 |
+
{
|
| 3005 |
+
"epoch": 32.25806451612903,
|
| 3006 |
+
"eval_loss": 0.12638460099697113,
|
| 3007 |
+
"eval_runtime": 101.1824,
|
| 3008 |
+
"eval_samples_per_second": 187.819,
|
| 3009 |
+
"eval_steps_per_second": 3.914,
|
| 3010 |
+
"step": 10000
|
| 3011 |
+
},
|
| 3012 |
+
{
|
| 3013 |
+
"epoch": 32.41935483870968,
|
| 3014 |
+
"grad_norm": 0.2691047489643097,
|
| 3015 |
+
"learning_rate": 7.033548387096774e-05,
|
| 3016 |
+
"loss": 0.1224,
|
| 3017 |
+
"step": 10050
|
| 3018 |
+
},
|
| 3019 |
+
{
|
| 3020 |
+
"epoch": 32.41935483870968,
|
| 3021 |
+
"eval_loss": 0.12849974632263184,
|
| 3022 |
+
"eval_runtime": 91.3103,
|
| 3023 |
+
"eval_samples_per_second": 208.126,
|
| 3024 |
+
"eval_steps_per_second": 4.337,
|
| 3025 |
+
"step": 10050
|
| 3026 |
+
},
|
| 3027 |
+
{
|
| 3028 |
+
"epoch": 32.58064516129032,
|
| 3029 |
+
"grad_norm": 0.30258217453956604,
|
| 3030 |
+
"learning_rate": 6.969032258064516e-05,
|
| 3031 |
+
"loss": 0.1239,
|
| 3032 |
+
"step": 10100
|
| 3033 |
+
},
|
| 3034 |
+
{
|
| 3035 |
+
"epoch": 32.58064516129032,
|
| 3036 |
+
"eval_loss": 0.12675543129444122,
|
| 3037 |
+
"eval_runtime": 87.3987,
|
| 3038 |
+
"eval_samples_per_second": 217.44,
|
| 3039 |
+
"eval_steps_per_second": 4.531,
|
| 3040 |
+
"step": 10100
|
| 3041 |
+
},
|
| 3042 |
+
{
|
| 3043 |
+
"epoch": 32.74193548387097,
|
| 3044 |
+
"grad_norm": 0.2675139009952545,
|
| 3045 |
+
"learning_rate": 6.904516129032258e-05,
|
| 3046 |
+
"loss": 0.1214,
|
| 3047 |
+
"step": 10150
|
| 3048 |
+
},
|
| 3049 |
+
{
|
| 3050 |
+
"epoch": 32.74193548387097,
|
| 3051 |
+
"eval_loss": 0.12617285549640656,
|
| 3052 |
+
"eval_runtime": 87.9466,
|
| 3053 |
+
"eval_samples_per_second": 216.086,
|
| 3054 |
+
"eval_steps_per_second": 4.503,
|
| 3055 |
+
"step": 10150
|
| 3056 |
+
},
|
| 3057 |
+
{
|
| 3058 |
+
"epoch": 32.903225806451616,
|
| 3059 |
+
"grad_norm": 0.3610474467277527,
|
| 3060 |
+
"learning_rate": 6.840000000000001e-05,
|
| 3061 |
+
"loss": 0.1205,
|
| 3062 |
+
"step": 10200
|
| 3063 |
+
},
|
| 3064 |
+
{
|
| 3065 |
+
"epoch": 32.903225806451616,
|
| 3066 |
+
"eval_loss": 0.12886326014995575,
|
| 3067 |
+
"eval_runtime": 88.5505,
|
| 3068 |
+
"eval_samples_per_second": 214.612,
|
| 3069 |
+
"eval_steps_per_second": 4.472,
|
| 3070 |
+
"step": 10200
|
| 3071 |
+
},
|
| 3072 |
+
{
|
| 3073 |
+
"epoch": 33.064516129032256,
|
| 3074 |
+
"grad_norm": 0.3488837480545044,
|
| 3075 |
+
"learning_rate": 6.775483870967742e-05,
|
| 3076 |
+
"loss": 0.1235,
|
| 3077 |
+
"step": 10250
|
| 3078 |
+
},
|
| 3079 |
+
{
|
| 3080 |
+
"epoch": 33.064516129032256,
|
| 3081 |
+
"eval_loss": 0.12522108852863312,
|
| 3082 |
+
"eval_runtime": 86.8524,
|
| 3083 |
+
"eval_samples_per_second": 218.808,
|
| 3084 |
+
"eval_steps_per_second": 4.559,
|
| 3085 |
+
"step": 10250
|
| 3086 |
+
},
|
| 3087 |
+
{
|
| 3088 |
+
"epoch": 33.225806451612904,
|
| 3089 |
+
"grad_norm": 0.31276750564575195,
|
| 3090 |
+
"learning_rate": 6.710967741935485e-05,
|
| 3091 |
+
"loss": 0.1205,
|
| 3092 |
+
"step": 10300
|
| 3093 |
+
},
|
| 3094 |
+
{
|
| 3095 |
+
"epoch": 33.225806451612904,
|
| 3096 |
+
"eval_loss": 0.12656815350055695,
|
| 3097 |
+
"eval_runtime": 88.7072,
|
| 3098 |
+
"eval_samples_per_second": 214.233,
|
| 3099 |
+
"eval_steps_per_second": 4.464,
|
| 3100 |
+
"step": 10300
|
| 3101 |
+
},
|
| 3102 |
+
{
|
| 3103 |
+
"epoch": 33.38709677419355,
|
| 3104 |
+
"grad_norm": 0.2674780488014221,
|
| 3105 |
+
"learning_rate": 6.646451612903226e-05,
|
| 3106 |
+
"loss": 0.1189,
|
| 3107 |
+
"step": 10350
|
| 3108 |
+
},
|
| 3109 |
+
{
|
| 3110 |
+
"epoch": 33.38709677419355,
|
| 3111 |
+
"eval_loss": 0.12716618180274963,
|
| 3112 |
+
"eval_runtime": 86.5042,
|
| 3113 |
+
"eval_samples_per_second": 219.689,
|
| 3114 |
+
"eval_steps_per_second": 4.578,
|
| 3115 |
+
"step": 10350
|
| 3116 |
+
},
|
| 3117 |
+
{
|
| 3118 |
+
"epoch": 33.54838709677419,
|
| 3119 |
+
"grad_norm": 0.35878920555114746,
|
| 3120 |
+
"learning_rate": 6.581935483870969e-05,
|
| 3121 |
+
"loss": 0.1187,
|
| 3122 |
+
"step": 10400
|
| 3123 |
+
},
|
| 3124 |
+
{
|
| 3125 |
+
"epoch": 33.54838709677419,
|
| 3126 |
+
"eval_loss": 0.12553632259368896,
|
| 3127 |
+
"eval_runtime": 89.3045,
|
| 3128 |
+
"eval_samples_per_second": 212.8,
|
| 3129 |
+
"eval_steps_per_second": 4.434,
|
| 3130 |
+
"step": 10400
|
| 3131 |
+
},
|
| 3132 |
+
{
|
| 3133 |
+
"epoch": 33.70967741935484,
|
| 3134 |
+
"grad_norm": 0.3341622054576874,
|
| 3135 |
+
"learning_rate": 6.51741935483871e-05,
|
| 3136 |
+
"loss": 0.1256,
|
| 3137 |
+
"step": 10450
|
| 3138 |
+
},
|
| 3139 |
+
{
|
| 3140 |
+
"epoch": 33.70967741935484,
|
| 3141 |
+
"eval_loss": 0.12624432146549225,
|
| 3142 |
+
"eval_runtime": 88.9284,
|
| 3143 |
+
"eval_samples_per_second": 213.7,
|
| 3144 |
+
"eval_steps_per_second": 4.453,
|
| 3145 |
+
"step": 10450
|
| 3146 |
+
},
|
| 3147 |
+
{
|
| 3148 |
+
"epoch": 33.87096774193548,
|
| 3149 |
+
"grad_norm": 0.30921441316604614,
|
| 3150 |
+
"learning_rate": 6.452903225806451e-05,
|
| 3151 |
+
"loss": 0.1226,
|
| 3152 |
+
"step": 10500
|
| 3153 |
+
},
|
| 3154 |
+
{
|
| 3155 |
+
"epoch": 33.87096774193548,
|
| 3156 |
+
"eval_loss": 0.12660150229930878,
|
| 3157 |
+
"eval_runtime": 89.8795,
|
| 3158 |
+
"eval_samples_per_second": 211.439,
|
| 3159 |
+
"eval_steps_per_second": 4.406,
|
| 3160 |
+
"step": 10500
|
| 3161 |
+
},
|
| 3162 |
+
{
|
| 3163 |
+
"epoch": 34.03225806451613,
|
| 3164 |
+
"grad_norm": 0.39118140935897827,
|
| 3165 |
+
"learning_rate": 6.388387096774194e-05,
|
| 3166 |
+
"loss": 0.1239,
|
| 3167 |
+
"step": 10550
|
| 3168 |
+
},
|
| 3169 |
+
{
|
| 3170 |
+
"epoch": 34.03225806451613,
|
| 3171 |
+
"eval_loss": 0.1243576630949974,
|
| 3172 |
+
"eval_runtime": 89.012,
|
| 3173 |
+
"eval_samples_per_second": 213.499,
|
| 3174 |
+
"eval_steps_per_second": 4.449,
|
| 3175 |
+
"step": 10550
|
| 3176 |
+
},
|
| 3177 |
+
{
|
| 3178 |
+
"epoch": 34.193548387096776,
|
| 3179 |
+
"grad_norm": 0.3129843771457672,
|
| 3180 |
+
"learning_rate": 6.323870967741936e-05,
|
| 3181 |
+
"loss": 0.1227,
|
| 3182 |
+
"step": 10600
|
| 3183 |
+
},
|
| 3184 |
+
{
|
| 3185 |
+
"epoch": 34.193548387096776,
|
| 3186 |
+
"eval_loss": 0.12571550905704498,
|
| 3187 |
+
"eval_runtime": 88.7287,
|
| 3188 |
+
"eval_samples_per_second": 214.181,
|
| 3189 |
+
"eval_steps_per_second": 4.463,
|
| 3190 |
+
"step": 10600
|
| 3191 |
+
},
|
| 3192 |
+
{
|
| 3193 |
+
"epoch": 34.354838709677416,
|
| 3194 |
+
"grad_norm": 0.3165799081325531,
|
| 3195 |
+
"learning_rate": 6.259354838709678e-05,
|
| 3196 |
+
"loss": 0.1176,
|
| 3197 |
+
"step": 10650
|
| 3198 |
+
},
|
| 3199 |
+
{
|
| 3200 |
+
"epoch": 34.354838709677416,
|
| 3201 |
+
"eval_loss": 0.12522710859775543,
|
| 3202 |
+
"eval_runtime": 87.961,
|
| 3203 |
+
"eval_samples_per_second": 216.05,
|
| 3204 |
+
"eval_steps_per_second": 4.502,
|
| 3205 |
+
"step": 10650
|
| 3206 |
+
},
|
| 3207 |
+
{
|
| 3208 |
+
"epoch": 34.516129032258064,
|
| 3209 |
+
"grad_norm": 0.30507832765579224,
|
| 3210 |
+
"learning_rate": 6.19483870967742e-05,
|
| 3211 |
+
"loss": 0.1224,
|
| 3212 |
+
"step": 10700
|
| 3213 |
+
},
|
| 3214 |
+
{
|
| 3215 |
+
"epoch": 34.516129032258064,
|
| 3216 |
+
"eval_loss": 0.1257481724023819,
|
| 3217 |
+
"eval_runtime": 88.8686,
|
| 3218 |
+
"eval_samples_per_second": 213.844,
|
| 3219 |
+
"eval_steps_per_second": 4.456,
|
| 3220 |
+
"step": 10700
|
| 3221 |
+
},
|
| 3222 |
+
{
|
| 3223 |
+
"epoch": 34.67741935483871,
|
| 3224 |
+
"grad_norm": 0.35299909114837646,
|
| 3225 |
+
"learning_rate": 6.130322580645161e-05,
|
| 3226 |
+
"loss": 0.1194,
|
| 3227 |
+
"step": 10750
|
| 3228 |
+
},
|
| 3229 |
+
{
|
| 3230 |
+
"epoch": 34.67741935483871,
|
| 3231 |
+
"eval_loss": 0.12496702373027802,
|
| 3232 |
+
"eval_runtime": 90.6297,
|
| 3233 |
+
"eval_samples_per_second": 209.689,
|
| 3234 |
+
"eval_steps_per_second": 4.369,
|
| 3235 |
+
"step": 10750
|
| 3236 |
+
},
|
| 3237 |
+
{
|
| 3238 |
+
"epoch": 34.83870967741935,
|
| 3239 |
+
"grad_norm": 0.29007554054260254,
|
| 3240 |
+
"learning_rate": 6.065806451612903e-05,
|
| 3241 |
+
"loss": 0.1187,
|
| 3242 |
+
"step": 10800
|
| 3243 |
+
},
|
| 3244 |
+
{
|
| 3245 |
+
"epoch": 34.83870967741935,
|
| 3246 |
+
"eval_loss": 0.12488405406475067,
|
| 3247 |
+
"eval_runtime": 88.9174,
|
| 3248 |
+
"eval_samples_per_second": 213.726,
|
| 3249 |
+
"eval_steps_per_second": 4.454,
|
| 3250 |
+
"step": 10800
|
| 3251 |
+
},
|
| 3252 |
+
{
|
| 3253 |
+
"epoch": 35.0,
|
| 3254 |
+
"grad_norm": 0.48845478892326355,
|
| 3255 |
+
"learning_rate": 6.001290322580645e-05,
|
| 3256 |
+
"loss": 0.1196,
|
| 3257 |
+
"step": 10850
|
| 3258 |
+
},
|
| 3259 |
+
{
|
| 3260 |
+
"epoch": 35.0,
|
| 3261 |
+
"eval_loss": 0.12530028820037842,
|
| 3262 |
+
"eval_runtime": 89.441,
|
| 3263 |
+
"eval_samples_per_second": 212.475,
|
| 3264 |
+
"eval_steps_per_second": 4.428,
|
| 3265 |
+
"step": 10850
|
| 3266 |
+
},
|
| 3267 |
+
{
|
| 3268 |
+
"epoch": 35.16129032258065,
|
| 3269 |
+
"grad_norm": 0.25860869884490967,
|
| 3270 |
+
"learning_rate": 5.936774193548388e-05,
|
| 3271 |
+
"loss": 0.1183,
|
| 3272 |
+
"step": 10900
|
| 3273 |
+
},
|
| 3274 |
+
{
|
| 3275 |
+
"epoch": 35.16129032258065,
|
| 3276 |
+
"eval_loss": 0.12282554060220718,
|
| 3277 |
+
"eval_runtime": 88.2534,
|
| 3278 |
+
"eval_samples_per_second": 215.334,
|
| 3279 |
+
"eval_steps_per_second": 4.487,
|
| 3280 |
+
"step": 10900
|
| 3281 |
+
},
|
| 3282 |
+
{
|
| 3283 |
+
"epoch": 35.32258064516129,
|
| 3284 |
+
"grad_norm": 0.330858051776886,
|
| 3285 |
+
"learning_rate": 5.87225806451613e-05,
|
| 3286 |
+
"loss": 0.1157,
|
| 3287 |
+
"step": 10950
|
| 3288 |
+
},
|
| 3289 |
+
{
|
| 3290 |
+
"epoch": 35.32258064516129,
|
| 3291 |
+
"eval_loss": 0.12600766122341156,
|
| 3292 |
+
"eval_runtime": 88.4405,
|
| 3293 |
+
"eval_samples_per_second": 214.879,
|
| 3294 |
+
"eval_steps_per_second": 4.478,
|
| 3295 |
+
"step": 10950
|
| 3296 |
+
},
|
| 3297 |
+
{
|
| 3298 |
+
"epoch": 35.483870967741936,
|
| 3299 |
+
"grad_norm": 0.2873861789703369,
|
| 3300 |
+
"learning_rate": 5.8077419354838716e-05,
|
| 3301 |
+
"loss": 0.1188,
|
| 3302 |
+
"step": 11000
|
| 3303 |
+
},
|
| 3304 |
+
{
|
| 3305 |
+
"epoch": 35.483870967741936,
|
| 3306 |
+
"eval_loss": 0.12375803291797638,
|
| 3307 |
+
"eval_runtime": 90.5675,
|
| 3308 |
+
"eval_samples_per_second": 209.832,
|
| 3309 |
+
"eval_steps_per_second": 4.372,
|
| 3310 |
+
"step": 11000
|
| 3311 |
+
},
|
| 3312 |
+
{
|
| 3313 |
+
"epoch": 35.645161290322584,
|
| 3314 |
+
"grad_norm": 0.3179507851600647,
|
| 3315 |
+
"learning_rate": 5.743225806451613e-05,
|
| 3316 |
+
"loss": 0.1166,
|
| 3317 |
+
"step": 11050
|
| 3318 |
+
},
|
| 3319 |
+
{
|
| 3320 |
+
"epoch": 35.645161290322584,
|
| 3321 |
+
"eval_loss": 0.12402182072401047,
|
| 3322 |
+
"eval_runtime": 89.7815,
|
| 3323 |
+
"eval_samples_per_second": 211.669,
|
| 3324 |
+
"eval_steps_per_second": 4.411,
|
| 3325 |
+
"step": 11050
|
| 3326 |
+
},
|
| 3327 |
+
{
|
| 3328 |
+
"epoch": 35.806451612903224,
|
| 3329 |
+
"grad_norm": 0.2623940110206604,
|
| 3330 |
+
"learning_rate": 5.678709677419355e-05,
|
| 3331 |
+
"loss": 0.116,
|
| 3332 |
+
"step": 11100
|
| 3333 |
+
},
|
| 3334 |
+
{
|
| 3335 |
+
"epoch": 35.806451612903224,
|
| 3336 |
+
"eval_loss": 0.12476334720849991,
|
| 3337 |
+
"eval_runtime": 89.8543,
|
| 3338 |
+
"eval_samples_per_second": 211.498,
|
| 3339 |
+
"eval_steps_per_second": 4.407,
|
| 3340 |
+
"step": 11100
|
| 3341 |
+
},
|
| 3342 |
+
{
|
| 3343 |
+
"epoch": 35.96774193548387,
|
| 3344 |
+
"grad_norm": 0.2534388601779938,
|
| 3345 |
+
"learning_rate": 5.614193548387097e-05,
|
| 3346 |
+
"loss": 0.1207,
|
| 3347 |
+
"step": 11150
|
| 3348 |
+
},
|
| 3349 |
+
{
|
| 3350 |
+
"epoch": 35.96774193548387,
|
| 3351 |
+
"eval_loss": 0.12570072710514069,
|
| 3352 |
+
"eval_runtime": 91.2008,
|
| 3353 |
+
"eval_samples_per_second": 208.375,
|
| 3354 |
+
"eval_steps_per_second": 4.342,
|
| 3355 |
+
"step": 11150
|
| 3356 |
+
},
|
| 3357 |
+
{
|
| 3358 |
+
"epoch": 36.12903225806452,
|
| 3359 |
+
"grad_norm": 0.30418768525123596,
|
| 3360 |
+
"learning_rate": 5.5496774193548386e-05,
|
| 3361 |
+
"loss": 0.1177,
|
| 3362 |
+
"step": 11200
|
| 3363 |
+
},
|
| 3364 |
+
{
|
| 3365 |
+
"epoch": 36.12903225806452,
|
| 3366 |
+
"eval_loss": 0.12448415905237198,
|
| 3367 |
+
"eval_runtime": 89.3046,
|
| 3368 |
+
"eval_samples_per_second": 212.8,
|
| 3369 |
+
"eval_steps_per_second": 4.434,
|
| 3370 |
+
"step": 11200
|
| 3371 |
+
},
|
| 3372 |
+
{
|
| 3373 |
+
"epoch": 36.29032258064516,
|
| 3374 |
+
"grad_norm": 0.290436327457428,
|
| 3375 |
+
"learning_rate": 5.485161290322581e-05,
|
| 3376 |
+
"loss": 0.1147,
|
| 3377 |
+
"step": 11250
|
| 3378 |
+
},
|
| 3379 |
+
{
|
| 3380 |
+
"epoch": 36.29032258064516,
|
| 3381 |
+
"eval_loss": 0.12383412569761276,
|
| 3382 |
+
"eval_runtime": 91.1386,
|
| 3383 |
+
"eval_samples_per_second": 208.517,
|
| 3384 |
+
"eval_steps_per_second": 4.345,
|
| 3385 |
+
"step": 11250
|
| 3386 |
+
},
|
| 3387 |
+
{
|
| 3388 |
+
"epoch": 36.45161290322581,
|
| 3389 |
+
"grad_norm": 0.3162536919116974,
|
| 3390 |
+
"learning_rate": 5.420645161290323e-05,
|
| 3391 |
+
"loss": 0.1154,
|
| 3392 |
+
"step": 11300
|
| 3393 |
+
},
|
| 3394 |
+
{
|
| 3395 |
+
"epoch": 36.45161290322581,
|
| 3396 |
+
"eval_loss": 0.12289289385080338,
|
| 3397 |
+
"eval_runtime": 90.6693,
|
| 3398 |
+
"eval_samples_per_second": 209.597,
|
| 3399 |
+
"eval_steps_per_second": 4.368,
|
| 3400 |
+
"step": 11300
|
| 3401 |
+
},
|
| 3402 |
+
{
|
| 3403 |
+
"epoch": 36.61290322580645,
|
| 3404 |
+
"grad_norm": 0.3101736903190613,
|
| 3405 |
+
"learning_rate": 5.356129032258065e-05,
|
| 3406 |
+
"loss": 0.1146,
|
| 3407 |
+
"step": 11350
|
| 3408 |
+
},
|
| 3409 |
+
{
|
| 3410 |
+
"epoch": 36.61290322580645,
|
| 3411 |
+
"eval_loss": 0.12234646826982498,
|
| 3412 |
+
"eval_runtime": 89.6047,
|
| 3413 |
+
"eval_samples_per_second": 212.087,
|
| 3414 |
+
"eval_steps_per_second": 4.419,
|
| 3415 |
+
"step": 11350
|
| 3416 |
+
},
|
| 3417 |
+
{
|
| 3418 |
+
"epoch": 36.774193548387096,
|
| 3419 |
+
"grad_norm": 0.32668620347976685,
|
| 3420 |
+
"learning_rate": 5.291612903225806e-05,
|
| 3421 |
+
"loss": 0.1196,
|
| 3422 |
+
"step": 11400
|
| 3423 |
+
},
|
| 3424 |
+
{
|
| 3425 |
+
"epoch": 36.774193548387096,
|
| 3426 |
+
"eval_loss": 0.12317313253879547,
|
| 3427 |
+
"eval_runtime": 88.2522,
|
| 3428 |
+
"eval_samples_per_second": 215.337,
|
| 3429 |
+
"eval_steps_per_second": 4.487,
|
| 3430 |
+
"step": 11400
|
| 3431 |
+
},
|
| 3432 |
+
{
|
| 3433 |
+
"epoch": 36.935483870967744,
|
| 3434 |
+
"grad_norm": 0.3477221429347992,
|
| 3435 |
+
"learning_rate": 5.227096774193548e-05,
|
| 3436 |
+
"loss": 0.1165,
|
| 3437 |
+
"step": 11450
|
| 3438 |
+
},
|
| 3439 |
+
{
|
| 3440 |
+
"epoch": 36.935483870967744,
|
| 3441 |
+
"eval_loss": 0.12321442365646362,
|
| 3442 |
+
"eval_runtime": 90.4924,
|
| 3443 |
+
"eval_samples_per_second": 210.007,
|
| 3444 |
+
"eval_steps_per_second": 4.376,
|
| 3445 |
+
"step": 11450
|
| 3446 |
+
},
|
| 3447 |
+
{
|
| 3448 |
+
"epoch": 37.096774193548384,
|
| 3449 |
+
"grad_norm": 0.3417079448699951,
|
| 3450 |
+
"learning_rate": 5.16258064516129e-05,
|
| 3451 |
+
"loss": 0.1193,
|
| 3452 |
+
"step": 11500
|
| 3453 |
+
},
|
| 3454 |
+
{
|
| 3455 |
+
"epoch": 37.096774193548384,
|
| 3456 |
+
"eval_loss": 0.12433473765850067,
|
| 3457 |
+
"eval_runtime": 87.8702,
|
| 3458 |
+
"eval_samples_per_second": 216.273,
|
| 3459 |
+
"eval_steps_per_second": 4.507,
|
| 3460 |
+
"step": 11500
|
| 3461 |
+
},
|
| 3462 |
+
{
|
| 3463 |
+
"epoch": 37.25806451612903,
|
| 3464 |
+
"grad_norm": 0.4154585301876068,
|
| 3465 |
+
"learning_rate": 5.098064516129033e-05,
|
| 3466 |
+
"loss": 0.1178,
|
| 3467 |
+
"step": 11550
|
| 3468 |
+
},
|
| 3469 |
+
{
|
| 3470 |
+
"epoch": 37.25806451612903,
|
| 3471 |
+
"eval_loss": 0.12312240153551102,
|
| 3472 |
+
"eval_runtime": 87.9067,
|
| 3473 |
+
"eval_samples_per_second": 216.184,
|
| 3474 |
+
"eval_steps_per_second": 4.505,
|
| 3475 |
+
"step": 11550
|
| 3476 |
+
},
|
| 3477 |
+
{
|
| 3478 |
+
"epoch": 37.41935483870968,
|
| 3479 |
+
"grad_norm": 0.3054625988006592,
|
| 3480 |
+
"learning_rate": 5.0335483870967747e-05,
|
| 3481 |
+
"loss": 0.1175,
|
| 3482 |
+
"step": 11600
|
| 3483 |
+
},
|
| 3484 |
+
{
|
| 3485 |
+
"epoch": 37.41935483870968,
|
| 3486 |
+
"eval_loss": 0.12180905044078827,
|
| 3487 |
+
"eval_runtime": 88.3639,
|
| 3488 |
+
"eval_samples_per_second": 215.065,
|
| 3489 |
+
"eval_steps_per_second": 4.481,
|
| 3490 |
+
"step": 11600
|
| 3491 |
+
},
|
| 3492 |
+
{
|
| 3493 |
+
"epoch": 37.58064516129032,
|
| 3494 |
+
"grad_norm": 0.28812453150749207,
|
| 3495 |
+
"learning_rate": 4.9690322580645166e-05,
|
| 3496 |
+
"loss": 0.117,
|
| 3497 |
+
"step": 11650
|
| 3498 |
+
},
|
| 3499 |
+
{
|
| 3500 |
+
"epoch": 37.58064516129032,
|
| 3501 |
+
"eval_loss": 0.12359043955802917,
|
| 3502 |
+
"eval_runtime": 88.3972,
|
| 3503 |
+
"eval_samples_per_second": 214.984,
|
| 3504 |
+
"eval_steps_per_second": 4.48,
|
| 3505 |
+
"step": 11650
|
| 3506 |
+
},
|
| 3507 |
+
{
|
| 3508 |
+
"epoch": 37.74193548387097,
|
| 3509 |
+
"grad_norm": 0.23818424344062805,
|
| 3510 |
+
"learning_rate": 4.9045161290322585e-05,
|
| 3511 |
+
"loss": 0.115,
|
| 3512 |
+
"step": 11700
|
| 3513 |
+
},
|
| 3514 |
+
{
|
| 3515 |
+
"epoch": 37.74193548387097,
|
| 3516 |
+
"eval_loss": 0.12239911407232285,
|
| 3517 |
+
"eval_runtime": 89.3047,
|
| 3518 |
+
"eval_samples_per_second": 212.8,
|
| 3519 |
+
"eval_steps_per_second": 4.434,
|
| 3520 |
+
"step": 11700
|
| 3521 |
+
},
|
| 3522 |
+
{
|
| 3523 |
+
"epoch": 37.903225806451616,
|
| 3524 |
+
"grad_norm": 0.28868499398231506,
|
| 3525 |
+
"learning_rate": 4.8400000000000004e-05,
|
| 3526 |
+
"loss": 0.1176,
|
| 3527 |
+
"step": 11750
|
| 3528 |
+
},
|
| 3529 |
+
{
|
| 3530 |
+
"epoch": 37.903225806451616,
|
| 3531 |
+
"eval_loss": 0.12191120535135269,
|
| 3532 |
+
"eval_runtime": 88.9578,
|
| 3533 |
+
"eval_samples_per_second": 213.629,
|
| 3534 |
+
"eval_steps_per_second": 4.452,
|
| 3535 |
+
"step": 11750
|
| 3536 |
+
},
|
| 3537 |
+
{
|
| 3538 |
+
"epoch": 38.064516129032256,
|
| 3539 |
+
"grad_norm": 0.33930733799934387,
|
| 3540 |
+
"learning_rate": 4.775483870967742e-05,
|
| 3541 |
+
"loss": 0.1145,
|
| 3542 |
+
"step": 11800
|
| 3543 |
+
},
|
| 3544 |
+
{
|
| 3545 |
+
"epoch": 38.064516129032256,
|
| 3546 |
+
"eval_loss": 0.12135030329227448,
|
| 3547 |
+
"eval_runtime": 90.281,
|
| 3548 |
+
"eval_samples_per_second": 210.498,
|
| 3549 |
+
"eval_steps_per_second": 4.386,
|
| 3550 |
+
"step": 11800
|
| 3551 |
+
},
|
| 3552 |
+
{
|
| 3553 |
+
"epoch": 38.225806451612904,
|
| 3554 |
+
"grad_norm": 0.3511495590209961,
|
| 3555 |
+
"learning_rate": 4.710967741935484e-05,
|
| 3556 |
+
"loss": 0.1161,
|
| 3557 |
+
"step": 11850
|
| 3558 |
+
},
|
| 3559 |
+
{
|
| 3560 |
+
"epoch": 38.225806451612904,
|
| 3561 |
+
"eval_loss": 0.12242971360683441,
|
| 3562 |
+
"eval_runtime": 89.1539,
|
| 3563 |
+
"eval_samples_per_second": 213.159,
|
| 3564 |
+
"eval_steps_per_second": 4.442,
|
| 3565 |
+
"step": 11850
|
| 3566 |
+
},
|
| 3567 |
+
{
|
| 3568 |
+
"epoch": 38.38709677419355,
|
| 3569 |
+
"grad_norm": 0.37273716926574707,
|
| 3570 |
+
"learning_rate": 4.646451612903226e-05,
|
| 3571 |
+
"loss": 0.1107,
|
| 3572 |
+
"step": 11900
|
| 3573 |
+
},
|
| 3574 |
+
{
|
| 3575 |
+
"epoch": 38.38709677419355,
|
| 3576 |
+
"eval_loss": 0.12210850417613983,
|
| 3577 |
+
"eval_runtime": 89.9424,
|
| 3578 |
+
"eval_samples_per_second": 211.291,
|
| 3579 |
+
"eval_steps_per_second": 4.403,
|
| 3580 |
+
"step": 11900
|
| 3581 |
+
},
|
| 3582 |
+
{
|
| 3583 |
+
"epoch": 38.54838709677419,
|
| 3584 |
+
"grad_norm": 0.2974016070365906,
|
| 3585 |
+
"learning_rate": 4.5819354838709674e-05,
|
| 3586 |
+
"loss": 0.1135,
|
| 3587 |
+
"step": 11950
|
| 3588 |
+
},
|
| 3589 |
+
{
|
| 3590 |
+
"epoch": 38.54838709677419,
|
| 3591 |
+
"eval_loss": 0.12212313711643219,
|
| 3592 |
+
"eval_runtime": 89.4298,
|
| 3593 |
+
"eval_samples_per_second": 212.502,
|
| 3594 |
+
"eval_steps_per_second": 4.428,
|
| 3595 |
+
"step": 11950
|
| 3596 |
+
},
|
| 3597 |
+
{
|
| 3598 |
+
"epoch": 38.70967741935484,
|
| 3599 |
+
"grad_norm": 0.32708504796028137,
|
| 3600 |
+
"learning_rate": 4.51741935483871e-05,
|
| 3601 |
+
"loss": 0.1167,
|
| 3602 |
+
"step": 12000
|
| 3603 |
+
},
|
| 3604 |
+
{
|
| 3605 |
+
"epoch": 38.70967741935484,
|
| 3606 |
+
"eval_loss": 0.12119368463754654,
|
| 3607 |
+
"eval_runtime": 88.7472,
|
| 3608 |
+
"eval_samples_per_second": 214.136,
|
| 3609 |
+
"eval_steps_per_second": 4.462,
|
| 3610 |
+
"step": 12000
|
| 3611 |
+
}
|
| 3612 |
+
],
|
| 3613 |
+
"logging_steps": 50,
|
| 3614 |
+
"max_steps": 15500,
|
| 3615 |
+
"num_input_tokens_seen": 0,
|
| 3616 |
+
"num_train_epochs": 50,
|
| 3617 |
+
"save_steps": 1000,
|
| 3618 |
+
"stateful_callbacks": {
|
| 3619 |
+
"TrainerControl": {
|
| 3620 |
+
"args": {
|
| 3621 |
+
"should_epoch_stop": false,
|
| 3622 |
+
"should_evaluate": false,
|
| 3623 |
+
"should_log": false,
|
| 3624 |
+
"should_save": true,
|
| 3625 |
+
"should_training_stop": false
|
| 3626 |
+
},
|
| 3627 |
+
"attributes": {}
|
| 3628 |
+
}
|
| 3629 |
+
},
|
| 3630 |
+
"total_flos": 1.2083464406274048e+17,
|
| 3631 |
+
"train_batch_size": 96,
|
| 3632 |
+
"trial_name": null,
|
| 3633 |
+
"trial_params": null
|
| 3634 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5d1f34eb3ddd6fe493ce04f41b2277a2e6325e8cc6221eb679f1d3cbc87664c1
|
| 3 |
+
size 15761
|