Training in progress, step 5000
Browse files- config.json +28 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state_0.pth +3 -0
- rng_state_1.pth +3 -0
- runs/Jun07_20-36-29_829f6f605e43/events.out.tfevents.1717792718.829f6f605e43.85.0 +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +37 -0
- tokenizer.json +0 -0
- tokenizer_config.json +53 -0
- trainer_state.json +1792 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"AlbertForMaskedLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0,
|
| 6 |
+
"bos_token_id": 2,
|
| 7 |
+
"classifier_dropout_prob": 0.1,
|
| 8 |
+
"embedding_size": 128,
|
| 9 |
+
"eos_token_id": 3,
|
| 10 |
+
"hidden_act": "gelu_new",
|
| 11 |
+
"hidden_dropout_prob": 0,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"inner_group_num": 1,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_norm_eps": 1e-12,
|
| 17 |
+
"max_position_embeddings": 32,
|
| 18 |
+
"model_type": "albert",
|
| 19 |
+
"num_attention_heads": 6,
|
| 20 |
+
"num_hidden_groups": 1,
|
| 21 |
+
"num_hidden_layers": 6,
|
| 22 |
+
"pad_token_id": 0,
|
| 23 |
+
"position_embedding_type": "absolute",
|
| 24 |
+
"torch_dtype": "float32",
|
| 25 |
+
"transformers_version": "4.41.2",
|
| 26 |
+
"type_vocab_size": 2,
|
| 27 |
+
"vocab_size": 30000
|
| 28 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8abaa42d3463b1fade9c9a8962318359c17b20036209deee3f1533c022f6ff45
|
| 3 |
+
size 44644496
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:82254252c8439c400690d9e912a9b41091500366183fc979cee1cef9b2d35ddb
|
| 3 |
+
size 11230198
|
rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6cb795a5cea0baa625c50007a6c9da09c6bbb5c16b560424070384a479e7d8a6
|
| 3 |
+
size 14512
|
rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5f19604377bd828eb366c68946ad997a4ff4d69beaeea93ee58915135768ec63
|
| 3 |
+
size 14512
|
runs/Jun07_20-36-29_829f6f605e43/events.out.tfevents.1717792718.829f6f605e43.85.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c5f5a58aa39d257803087286f6970f617b220fb828688a142a7a6f6b2ebe0a9a
|
| 3 |
+
size 57851
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fe25b65e375d09c26af85ce4a53e5909be64fae7753341d624d8971238990a51
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": {
|
| 3 |
+
"content": "[CLS]",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"mask_token": {
|
| 10 |
+
"content": "[MASK]",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "[PAD]",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"sep_token": {
|
| 24 |
+
"content": "[SEP]",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"unk_token": {
|
| 31 |
+
"content": "[UNK]",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
}
|
| 37 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "[SEP]",
|
| 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 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": true,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"mask_token": "[MASK]",
|
| 47 |
+
"model_max_length": 8192,
|
| 48 |
+
"pad_token": "[PAD]",
|
| 49 |
+
"sep_token": "[SEP]",
|
| 50 |
+
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 51 |
+
"unk_token": "[UNK]",
|
| 52 |
+
"use_fast_tokenizer": true
|
| 53 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1792 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.0904895484571532,
|
| 5 |
+
"eval_steps": 5000,
|
| 6 |
+
"global_step": 5000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0003619581938286128,
|
| 13 |
+
"grad_norm": 1.063517451286316,
|
| 14 |
+
"learning_rate": 1.0858418290402365e-06,
|
| 15 |
+
"loss": 10.3292,
|
| 16 |
+
"step": 20
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0007239163876572256,
|
| 20 |
+
"grad_norm": 0.7639716267585754,
|
| 21 |
+
"learning_rate": 2.171683658080473e-06,
|
| 22 |
+
"loss": 10.3095,
|
| 23 |
+
"step": 40
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0010858745814858383,
|
| 27 |
+
"grad_norm": 0.5479791760444641,
|
| 28 |
+
"learning_rate": 3.2575254871207094e-06,
|
| 29 |
+
"loss": 10.2858,
|
| 30 |
+
"step": 60
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0014478327753144511,
|
| 34 |
+
"grad_norm": 0.5541791319847107,
|
| 35 |
+
"learning_rate": 4.343367316160946e-06,
|
| 36 |
+
"loss": 10.2636,
|
| 37 |
+
"step": 80
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.001809790969143064,
|
| 41 |
+
"grad_norm": 0.5530744791030884,
|
| 42 |
+
"learning_rate": 5.429209145201182e-06,
|
| 43 |
+
"loss": 10.2491,
|
| 44 |
+
"step": 100
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.0021717491629716767,
|
| 48 |
+
"grad_norm": 0.5137726068496704,
|
| 49 |
+
"learning_rate": 6.515050974241419e-06,
|
| 50 |
+
"loss": 10.2343,
|
| 51 |
+
"step": 120
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.0025337073568002895,
|
| 55 |
+
"grad_norm": 0.512411892414093,
|
| 56 |
+
"learning_rate": 7.600892803281655e-06,
|
| 57 |
+
"loss": 10.2178,
|
| 58 |
+
"step": 140
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.0028956655506289022,
|
| 62 |
+
"grad_norm": 0.5807462334632874,
|
| 63 |
+
"learning_rate": 8.686734632321892e-06,
|
| 64 |
+
"loss": 10.1966,
|
| 65 |
+
"step": 160
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.003257623744457515,
|
| 69 |
+
"grad_norm": 0.7059533596038818,
|
| 70 |
+
"learning_rate": 9.772576461362129e-06,
|
| 71 |
+
"loss": 10.1748,
|
| 72 |
+
"step": 180
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.003619581938286128,
|
| 76 |
+
"grad_norm": 1.3133445978164673,
|
| 77 |
+
"learning_rate": 1.0858418290402365e-05,
|
| 78 |
+
"loss": 10.1451,
|
| 79 |
+
"step": 200
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.003981540132114741,
|
| 83 |
+
"grad_norm": 1.6613627672195435,
|
| 84 |
+
"learning_rate": 1.1944260119442602e-05,
|
| 85 |
+
"loss": 10.1148,
|
| 86 |
+
"step": 220
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.004343498325943353,
|
| 90 |
+
"grad_norm": 1.7141027450561523,
|
| 91 |
+
"learning_rate": 1.3030101948482837e-05,
|
| 92 |
+
"loss": 10.0838,
|
| 93 |
+
"step": 240
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.004705456519771966,
|
| 97 |
+
"grad_norm": 2.0595993995666504,
|
| 98 |
+
"learning_rate": 1.4115943777523073e-05,
|
| 99 |
+
"loss": 10.0519,
|
| 100 |
+
"step": 260
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.005067414713600579,
|
| 104 |
+
"grad_norm": 2.2947471141815186,
|
| 105 |
+
"learning_rate": 1.520178560656331e-05,
|
| 106 |
+
"loss": 10.0166,
|
| 107 |
+
"step": 280
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.005429372907429192,
|
| 111 |
+
"grad_norm": 2.2053394317626953,
|
| 112 |
+
"learning_rate": 1.628762743560355e-05,
|
| 113 |
+
"loss": 9.9809,
|
| 114 |
+
"step": 300
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.0057913311012578045,
|
| 118 |
+
"grad_norm": 2.2479448318481445,
|
| 119 |
+
"learning_rate": 1.7373469264643783e-05,
|
| 120 |
+
"loss": 9.9443,
|
| 121 |
+
"step": 320
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.006153289295086417,
|
| 125 |
+
"grad_norm": 1.9384366273880005,
|
| 126 |
+
"learning_rate": 1.845931109368402e-05,
|
| 127 |
+
"loss": 9.9023,
|
| 128 |
+
"step": 340
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.00651524748891503,
|
| 132 |
+
"grad_norm": 1.7514017820358276,
|
| 133 |
+
"learning_rate": 1.9545152922724258e-05,
|
| 134 |
+
"loss": 9.8601,
|
| 135 |
+
"step": 360
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.006877205682743643,
|
| 139 |
+
"grad_norm": 1.5420582294464111,
|
| 140 |
+
"learning_rate": 2.0630994751764492e-05,
|
| 141 |
+
"loss": 9.8214,
|
| 142 |
+
"step": 380
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.007239163876572256,
|
| 146 |
+
"grad_norm": 1.223652958869934,
|
| 147 |
+
"learning_rate": 2.171683658080473e-05,
|
| 148 |
+
"loss": 9.7768,
|
| 149 |
+
"step": 400
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.007601122070400868,
|
| 153 |
+
"grad_norm": 1.058700442314148,
|
| 154 |
+
"learning_rate": 2.2802678409844966e-05,
|
| 155 |
+
"loss": 9.7302,
|
| 156 |
+
"step": 420
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.007963080264229481,
|
| 160 |
+
"grad_norm": 1.0271687507629395,
|
| 161 |
+
"learning_rate": 2.3888520238885204e-05,
|
| 162 |
+
"loss": 9.6882,
|
| 163 |
+
"step": 440
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.008325038458058095,
|
| 167 |
+
"grad_norm": 0.9583302140235901,
|
| 168 |
+
"learning_rate": 2.4974362067925438e-05,
|
| 169 |
+
"loss": 9.6446,
|
| 170 |
+
"step": 460
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.008686996651886707,
|
| 174 |
+
"grad_norm": 0.8648065328598022,
|
| 175 |
+
"learning_rate": 2.6060203896965675e-05,
|
| 176 |
+
"loss": 9.6003,
|
| 177 |
+
"step": 480
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.00904895484571532,
|
| 181 |
+
"grad_norm": 0.7204363346099854,
|
| 182 |
+
"learning_rate": 2.7146045726005912e-05,
|
| 183 |
+
"loss": 9.5586,
|
| 184 |
+
"step": 500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.009410913039543932,
|
| 188 |
+
"grad_norm": 0.8680516481399536,
|
| 189 |
+
"learning_rate": 2.8231887555046146e-05,
|
| 190 |
+
"loss": 9.5111,
|
| 191 |
+
"step": 520
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.009772871233372546,
|
| 195 |
+
"grad_norm": 0.6860862374305725,
|
| 196 |
+
"learning_rate": 2.9317729384086387e-05,
|
| 197 |
+
"loss": 9.4771,
|
| 198 |
+
"step": 540
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.010134829427201158,
|
| 202 |
+
"grad_norm": 0.711931049823761,
|
| 203 |
+
"learning_rate": 3.040357121312662e-05,
|
| 204 |
+
"loss": 9.4369,
|
| 205 |
+
"step": 560
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.010496787621029771,
|
| 209 |
+
"grad_norm": 0.7360084652900696,
|
| 210 |
+
"learning_rate": 3.148941304216686e-05,
|
| 211 |
+
"loss": 9.3904,
|
| 212 |
+
"step": 580
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.010858745814858383,
|
| 216 |
+
"grad_norm": 0.7077947854995728,
|
| 217 |
+
"learning_rate": 3.25752548712071e-05,
|
| 218 |
+
"loss": 9.3552,
|
| 219 |
+
"step": 600
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.011220704008686997,
|
| 223 |
+
"grad_norm": 0.562315821647644,
|
| 224 |
+
"learning_rate": 3.366109670024733e-05,
|
| 225 |
+
"loss": 9.3164,
|
| 226 |
+
"step": 620
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.011582662202515609,
|
| 230 |
+
"grad_norm": 0.6069725751876831,
|
| 231 |
+
"learning_rate": 3.4746938529287566e-05,
|
| 232 |
+
"loss": 9.2838,
|
| 233 |
+
"step": 640
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.011944620396344223,
|
| 237 |
+
"grad_norm": 0.6498883962631226,
|
| 238 |
+
"learning_rate": 3.583278035832781e-05,
|
| 239 |
+
"loss": 9.2487,
|
| 240 |
+
"step": 660
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.012306578590172835,
|
| 244 |
+
"grad_norm": 0.615738034248352,
|
| 245 |
+
"learning_rate": 3.691862218736804e-05,
|
| 246 |
+
"loss": 9.2222,
|
| 247 |
+
"step": 680
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.012668536784001448,
|
| 251 |
+
"grad_norm": 0.5743957757949829,
|
| 252 |
+
"learning_rate": 3.8004464016408275e-05,
|
| 253 |
+
"loss": 9.1898,
|
| 254 |
+
"step": 700
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.01303049497783006,
|
| 258 |
+
"grad_norm": 0.5580448508262634,
|
| 259 |
+
"learning_rate": 3.9090305845448516e-05,
|
| 260 |
+
"loss": 9.1659,
|
| 261 |
+
"step": 720
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.013392453171658674,
|
| 265 |
+
"grad_norm": 0.5862032771110535,
|
| 266 |
+
"learning_rate": 4.017614767448875e-05,
|
| 267 |
+
"loss": 9.1376,
|
| 268 |
+
"step": 740
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.013754411365487286,
|
| 272 |
+
"grad_norm": 0.650966465473175,
|
| 273 |
+
"learning_rate": 4.1261989503528983e-05,
|
| 274 |
+
"loss": 9.1042,
|
| 275 |
+
"step": 760
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.0141163695593159,
|
| 279 |
+
"grad_norm": 0.6448431015014648,
|
| 280 |
+
"learning_rate": 4.2347831332569224e-05,
|
| 281 |
+
"loss": 9.0836,
|
| 282 |
+
"step": 780
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.014478327753144511,
|
| 286 |
+
"grad_norm": 0.5597690939903259,
|
| 287 |
+
"learning_rate": 4.343367316160946e-05,
|
| 288 |
+
"loss": 9.0728,
|
| 289 |
+
"step": 800
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.014840285946973125,
|
| 293 |
+
"grad_norm": 0.5503562092781067,
|
| 294 |
+
"learning_rate": 4.451951499064969e-05,
|
| 295 |
+
"loss": 9.0544,
|
| 296 |
+
"step": 820
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.015202244140801737,
|
| 300 |
+
"grad_norm": 0.5494912266731262,
|
| 301 |
+
"learning_rate": 4.560535681968993e-05,
|
| 302 |
+
"loss": 9.0229,
|
| 303 |
+
"step": 840
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.01556420233463035,
|
| 307 |
+
"grad_norm": 0.5567879676818848,
|
| 308 |
+
"learning_rate": 4.6691198648730167e-05,
|
| 309 |
+
"loss": 9.016,
|
| 310 |
+
"step": 860
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.015926160528458962,
|
| 314 |
+
"grad_norm": 0.5893707275390625,
|
| 315 |
+
"learning_rate": 4.777704047777041e-05,
|
| 316 |
+
"loss": 8.9989,
|
| 317 |
+
"step": 880
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.016288118722287574,
|
| 321 |
+
"grad_norm": 0.5258359909057617,
|
| 322 |
+
"learning_rate": 4.886288230681064e-05,
|
| 323 |
+
"loss": 8.9894,
|
| 324 |
+
"step": 900
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.01665007691611619,
|
| 328 |
+
"grad_norm": 0.5892924070358276,
|
| 329 |
+
"learning_rate": 4.9948724135850875e-05,
|
| 330 |
+
"loss": 8.9762,
|
| 331 |
+
"step": 920
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.0170120351099448,
|
| 335 |
+
"grad_norm": 0.6365529894828796,
|
| 336 |
+
"learning_rate": 5.1034565964891116e-05,
|
| 337 |
+
"loss": 8.9545,
|
| 338 |
+
"step": 940
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.017373993303773413,
|
| 342 |
+
"grad_norm": 0.5396739840507507,
|
| 343 |
+
"learning_rate": 5.212040779393135e-05,
|
| 344 |
+
"loss": 8.9387,
|
| 345 |
+
"step": 960
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.017735951497602025,
|
| 349 |
+
"grad_norm": 0.5578063130378723,
|
| 350 |
+
"learning_rate": 5.3206249622971584e-05,
|
| 351 |
+
"loss": 8.9452,
|
| 352 |
+
"step": 980
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.01809790969143064,
|
| 356 |
+
"grad_norm": 0.49258968234062195,
|
| 357 |
+
"learning_rate": 5.4292091452011824e-05,
|
| 358 |
+
"loss": 8.9329,
|
| 359 |
+
"step": 1000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.018459867885259253,
|
| 363 |
+
"grad_norm": 0.5692590475082397,
|
| 364 |
+
"learning_rate": 5.537793328105206e-05,
|
| 365 |
+
"loss": 8.9197,
|
| 366 |
+
"step": 1020
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.018821826079087865,
|
| 370 |
+
"grad_norm": 0.5728791356086731,
|
| 371 |
+
"learning_rate": 5.646377511009229e-05,
|
| 372 |
+
"loss": 8.915,
|
| 373 |
+
"step": 1040
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.019183784272916476,
|
| 377 |
+
"grad_norm": 0.4755489230155945,
|
| 378 |
+
"learning_rate": 5.754961693913253e-05,
|
| 379 |
+
"loss": 8.8982,
|
| 380 |
+
"step": 1060
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.019545742466745092,
|
| 384 |
+
"grad_norm": 0.5339570641517639,
|
| 385 |
+
"learning_rate": 5.8635458768172773e-05,
|
| 386 |
+
"loss": 8.896,
|
| 387 |
+
"step": 1080
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.019907700660573704,
|
| 391 |
+
"grad_norm": 0.58181232213974,
|
| 392 |
+
"learning_rate": 5.9721300597213e-05,
|
| 393 |
+
"loss": 8.8863,
|
| 394 |
+
"step": 1100
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.020269658854402316,
|
| 398 |
+
"grad_norm": 0.5068308115005493,
|
| 399 |
+
"learning_rate": 6.080714242625324e-05,
|
| 400 |
+
"loss": 8.8815,
|
| 401 |
+
"step": 1120
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.020631617048230928,
|
| 405 |
+
"grad_norm": 0.5282208919525146,
|
| 406 |
+
"learning_rate": 6.189298425529349e-05,
|
| 407 |
+
"loss": 8.8762,
|
| 408 |
+
"step": 1140
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.020993575242059543,
|
| 412 |
+
"grad_norm": 0.6251116991043091,
|
| 413 |
+
"learning_rate": 6.297882608433372e-05,
|
| 414 |
+
"loss": 8.875,
|
| 415 |
+
"step": 1160
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.021355533435888155,
|
| 419 |
+
"grad_norm": 0.7021293044090271,
|
| 420 |
+
"learning_rate": 6.406466791337396e-05,
|
| 421 |
+
"loss": 8.8549,
|
| 422 |
+
"step": 1180
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.021717491629716767,
|
| 426 |
+
"grad_norm": 0.5882667303085327,
|
| 427 |
+
"learning_rate": 6.51505097424142e-05,
|
| 428 |
+
"loss": 8.8602,
|
| 429 |
+
"step": 1200
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.022079449823545382,
|
| 433 |
+
"grad_norm": 0.552535891532898,
|
| 434 |
+
"learning_rate": 6.623635157145442e-05,
|
| 435 |
+
"loss": 8.8478,
|
| 436 |
+
"step": 1220
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.022441408017373994,
|
| 440 |
+
"grad_norm": 0.5536801218986511,
|
| 441 |
+
"learning_rate": 6.732219340049467e-05,
|
| 442 |
+
"loss": 8.8465,
|
| 443 |
+
"step": 1240
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.022803366211202606,
|
| 447 |
+
"grad_norm": 0.7319750189781189,
|
| 448 |
+
"learning_rate": 6.84080352295349e-05,
|
| 449 |
+
"loss": 8.8456,
|
| 450 |
+
"step": 1260
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.023165324405031218,
|
| 454 |
+
"grad_norm": 0.6104538440704346,
|
| 455 |
+
"learning_rate": 6.949387705857513e-05,
|
| 456 |
+
"loss": 8.8237,
|
| 457 |
+
"step": 1280
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.023527282598859833,
|
| 461 |
+
"grad_norm": 0.596066415309906,
|
| 462 |
+
"learning_rate": 7.057971888761537e-05,
|
| 463 |
+
"loss": 8.8247,
|
| 464 |
+
"step": 1300
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.023889240792688445,
|
| 468 |
+
"grad_norm": 0.516325056552887,
|
| 469 |
+
"learning_rate": 7.166556071665561e-05,
|
| 470 |
+
"loss": 8.8194,
|
| 471 |
+
"step": 1320
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.024251198986517057,
|
| 475 |
+
"grad_norm": 0.5828744173049927,
|
| 476 |
+
"learning_rate": 7.275140254569584e-05,
|
| 477 |
+
"loss": 8.8158,
|
| 478 |
+
"step": 1340
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.02461315718034567,
|
| 482 |
+
"grad_norm": 0.6400988101959229,
|
| 483 |
+
"learning_rate": 7.383724437473608e-05,
|
| 484 |
+
"loss": 8.797,
|
| 485 |
+
"step": 1360
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.024975115374174284,
|
| 489 |
+
"grad_norm": 0.6185368895530701,
|
| 490 |
+
"learning_rate": 7.492308620377632e-05,
|
| 491 |
+
"loss": 8.798,
|
| 492 |
+
"step": 1380
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.025337073568002896,
|
| 496 |
+
"grad_norm": 0.5269763469696045,
|
| 497 |
+
"learning_rate": 7.600892803281655e-05,
|
| 498 |
+
"loss": 8.7835,
|
| 499 |
+
"step": 1400
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.025699031761831508,
|
| 503 |
+
"grad_norm": 0.5572532415390015,
|
| 504 |
+
"learning_rate": 7.709476986185679e-05,
|
| 505 |
+
"loss": 8.7743,
|
| 506 |
+
"step": 1420
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.02606098995566012,
|
| 510 |
+
"grad_norm": 0.5819870233535767,
|
| 511 |
+
"learning_rate": 7.818061169089703e-05,
|
| 512 |
+
"loss": 8.7728,
|
| 513 |
+
"step": 1440
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.026422948149488736,
|
| 517 |
+
"grad_norm": 0.724423348903656,
|
| 518 |
+
"learning_rate": 7.926645351993726e-05,
|
| 519 |
+
"loss": 8.7537,
|
| 520 |
+
"step": 1460
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.026784906343317347,
|
| 524 |
+
"grad_norm": 0.729290246963501,
|
| 525 |
+
"learning_rate": 8.03522953489775e-05,
|
| 526 |
+
"loss": 8.7566,
|
| 527 |
+
"step": 1480
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.02714686453714596,
|
| 531 |
+
"grad_norm": 0.6978908777236938,
|
| 532 |
+
"learning_rate": 8.143813717801774e-05,
|
| 533 |
+
"loss": 8.7412,
|
| 534 |
+
"step": 1500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.02750882273097457,
|
| 538 |
+
"grad_norm": 0.9260867238044739,
|
| 539 |
+
"learning_rate": 8.252397900705797e-05,
|
| 540 |
+
"loss": 8.7469,
|
| 541 |
+
"step": 1520
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.027870780924803187,
|
| 545 |
+
"grad_norm": 0.8353539109230042,
|
| 546 |
+
"learning_rate": 8.360982083609821e-05,
|
| 547 |
+
"loss": 8.726,
|
| 548 |
+
"step": 1540
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.0282327391186318,
|
| 552 |
+
"grad_norm": 0.6905779242515564,
|
| 553 |
+
"learning_rate": 8.469566266513845e-05,
|
| 554 |
+
"loss": 8.7128,
|
| 555 |
+
"step": 1560
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.02859469731246041,
|
| 559 |
+
"grad_norm": 0.6228475570678711,
|
| 560 |
+
"learning_rate": 8.578150449417868e-05,
|
| 561 |
+
"loss": 8.7487,
|
| 562 |
+
"step": 1580
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.028956655506289022,
|
| 566 |
+
"grad_norm": 0.9750573635101318,
|
| 567 |
+
"learning_rate": 8.686734632321892e-05,
|
| 568 |
+
"loss": 8.7054,
|
| 569 |
+
"step": 1600
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.029318613700117638,
|
| 573 |
+
"grad_norm": 0.874622106552124,
|
| 574 |
+
"learning_rate": 8.795318815225916e-05,
|
| 575 |
+
"loss": 8.715,
|
| 576 |
+
"step": 1620
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.02968057189394625,
|
| 580 |
+
"grad_norm": 0.7310810685157776,
|
| 581 |
+
"learning_rate": 8.903902998129938e-05,
|
| 582 |
+
"loss": 8.6835,
|
| 583 |
+
"step": 1640
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.03004253008777486,
|
| 587 |
+
"grad_norm": 0.7120797038078308,
|
| 588 |
+
"learning_rate": 9.012487181033962e-05,
|
| 589 |
+
"loss": 8.6734,
|
| 590 |
+
"step": 1660
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.030404488281603474,
|
| 594 |
+
"grad_norm": 1.0088528394699097,
|
| 595 |
+
"learning_rate": 9.121071363937987e-05,
|
| 596 |
+
"loss": 8.6657,
|
| 597 |
+
"step": 1680
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.03076644647543209,
|
| 601 |
+
"grad_norm": 1.4212431907653809,
|
| 602 |
+
"learning_rate": 9.229655546842009e-05,
|
| 603 |
+
"loss": 8.6429,
|
| 604 |
+
"step": 1700
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.0311284046692607,
|
| 608 |
+
"grad_norm": 1.2623440027236938,
|
| 609 |
+
"learning_rate": 9.338239729746033e-05,
|
| 610 |
+
"loss": 8.6424,
|
| 611 |
+
"step": 1720
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.031490362863089316,
|
| 615 |
+
"grad_norm": 1.190894365310669,
|
| 616 |
+
"learning_rate": 9.446823912650057e-05,
|
| 617 |
+
"loss": 8.616,
|
| 618 |
+
"step": 1740
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.031852321056917925,
|
| 622 |
+
"grad_norm": 0.8714050054550171,
|
| 623 |
+
"learning_rate": 9.555408095554081e-05,
|
| 624 |
+
"loss": 8.6261,
|
| 625 |
+
"step": 1760
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.03221427925074654,
|
| 629 |
+
"grad_norm": 0.7428932189941406,
|
| 630 |
+
"learning_rate": 9.663992278458104e-05,
|
| 631 |
+
"loss": 8.6054,
|
| 632 |
+
"step": 1780
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.03257623744457515,
|
| 636 |
+
"grad_norm": 0.8327052593231201,
|
| 637 |
+
"learning_rate": 9.772576461362128e-05,
|
| 638 |
+
"loss": 8.6072,
|
| 639 |
+
"step": 1800
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.032938195638403764,
|
| 643 |
+
"grad_norm": 0.7944102883338928,
|
| 644 |
+
"learning_rate": 9.881160644266152e-05,
|
| 645 |
+
"loss": 8.5757,
|
| 646 |
+
"step": 1820
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.03330015383223238,
|
| 650 |
+
"grad_norm": 0.929286003112793,
|
| 651 |
+
"learning_rate": 9.989744827170175e-05,
|
| 652 |
+
"loss": 8.5651,
|
| 653 |
+
"step": 1840
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.03366211202606099,
|
| 657 |
+
"grad_norm": 0.9866963028907776,
|
| 658 |
+
"learning_rate": 0.00010098329010074199,
|
| 659 |
+
"loss": 8.55,
|
| 660 |
+
"step": 1860
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.0340240702198896,
|
| 664 |
+
"grad_norm": 1.5684864521026611,
|
| 665 |
+
"learning_rate": 0.00010206913192978223,
|
| 666 |
+
"loss": 8.5508,
|
| 667 |
+
"step": 1880
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.03438602841371822,
|
| 671 |
+
"grad_norm": 1.268312692642212,
|
| 672 |
+
"learning_rate": 0.00010315497375882246,
|
| 673 |
+
"loss": 8.5303,
|
| 674 |
+
"step": 1900
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.03474798660754683,
|
| 678 |
+
"grad_norm": 1.384734034538269,
|
| 679 |
+
"learning_rate": 0.0001042408155878627,
|
| 680 |
+
"loss": 8.526,
|
| 681 |
+
"step": 1920
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.03510994480137544,
|
| 685 |
+
"grad_norm": 1.1619597673416138,
|
| 686 |
+
"learning_rate": 0.00010532665741690294,
|
| 687 |
+
"loss": 8.5083,
|
| 688 |
+
"step": 1940
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.03547190299520405,
|
| 692 |
+
"grad_norm": 1.4747376441955566,
|
| 693 |
+
"learning_rate": 0.00010641249924594317,
|
| 694 |
+
"loss": 8.492,
|
| 695 |
+
"step": 1960
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.035833861189032666,
|
| 699 |
+
"grad_norm": 1.1740059852600098,
|
| 700 |
+
"learning_rate": 0.00010749834107498341,
|
| 701 |
+
"loss": 8.4921,
|
| 702 |
+
"step": 1980
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.03619581938286128,
|
| 706 |
+
"grad_norm": 1.6106988191604614,
|
| 707 |
+
"learning_rate": 0.00010858418290402365,
|
| 708 |
+
"loss": 8.4829,
|
| 709 |
+
"step": 2000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.03655777757668989,
|
| 713 |
+
"grad_norm": 1.2826563119888306,
|
| 714 |
+
"learning_rate": 0.00010967002473306388,
|
| 715 |
+
"loss": 8.4542,
|
| 716 |
+
"step": 2020
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.036919735770518505,
|
| 720 |
+
"grad_norm": 1.4442939758300781,
|
| 721 |
+
"learning_rate": 0.00011075586656210412,
|
| 722 |
+
"loss": 8.4472,
|
| 723 |
+
"step": 2040
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.03728169396434712,
|
| 727 |
+
"grad_norm": 1.3481926918029785,
|
| 728 |
+
"learning_rate": 0.00011184170839114436,
|
| 729 |
+
"loss": 8.4245,
|
| 730 |
+
"step": 2060
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.03764365215817573,
|
| 734 |
+
"grad_norm": 1.377825379371643,
|
| 735 |
+
"learning_rate": 0.00011292755022018458,
|
| 736 |
+
"loss": 8.4233,
|
| 737 |
+
"step": 2080
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.038005610352004345,
|
| 741 |
+
"grad_norm": 1.656119465827942,
|
| 742 |
+
"learning_rate": 0.00011401339204922482,
|
| 743 |
+
"loss": 8.4225,
|
| 744 |
+
"step": 2100
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.03836756854583295,
|
| 748 |
+
"grad_norm": 1.5955251455307007,
|
| 749 |
+
"learning_rate": 0.00011509923387826507,
|
| 750 |
+
"loss": 8.4216,
|
| 751 |
+
"step": 2120
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.03872952673966157,
|
| 755 |
+
"grad_norm": 1.6190309524536133,
|
| 756 |
+
"learning_rate": 0.0001161850757073053,
|
| 757 |
+
"loss": 8.4182,
|
| 758 |
+
"step": 2140
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.039091484933490184,
|
| 762 |
+
"grad_norm": 1.4823400974273682,
|
| 763 |
+
"learning_rate": 0.00011727091753634555,
|
| 764 |
+
"loss": 8.3738,
|
| 765 |
+
"step": 2160
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.03945344312731879,
|
| 769 |
+
"grad_norm": 4.104274749755859,
|
| 770 |
+
"learning_rate": 0.00011835675936538576,
|
| 771 |
+
"loss": 8.3636,
|
| 772 |
+
"step": 2180
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.03981540132114741,
|
| 776 |
+
"grad_norm": 1.6707680225372314,
|
| 777 |
+
"learning_rate": 0.000119442601194426,
|
| 778 |
+
"loss": 8.3745,
|
| 779 |
+
"step": 2200
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.04017735951497602,
|
| 783 |
+
"grad_norm": 1.1206501722335815,
|
| 784 |
+
"learning_rate": 0.00012052844302346624,
|
| 785 |
+
"loss": 8.353,
|
| 786 |
+
"step": 2220
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.04053931770880463,
|
| 790 |
+
"grad_norm": 2.2229607105255127,
|
| 791 |
+
"learning_rate": 0.00012161428485250648,
|
| 792 |
+
"loss": 8.3539,
|
| 793 |
+
"step": 2240
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.04090127590263325,
|
| 797 |
+
"grad_norm": 1.9011199474334717,
|
| 798 |
+
"learning_rate": 0.00012270012668154674,
|
| 799 |
+
"loss": 8.3293,
|
| 800 |
+
"step": 2260
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.041263234096461855,
|
| 804 |
+
"grad_norm": 2.2467918395996094,
|
| 805 |
+
"learning_rate": 0.00012378596851058698,
|
| 806 |
+
"loss": 8.326,
|
| 807 |
+
"step": 2280
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.04162519229029047,
|
| 811 |
+
"grad_norm": 1.530720829963684,
|
| 812 |
+
"learning_rate": 0.0001248718103396272,
|
| 813 |
+
"loss": 8.3197,
|
| 814 |
+
"step": 2300
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.041987150484119086,
|
| 818 |
+
"grad_norm": 2.6763076782226562,
|
| 819 |
+
"learning_rate": 0.00012595765216866743,
|
| 820 |
+
"loss": 8.3004,
|
| 821 |
+
"step": 2320
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.042349108677947694,
|
| 825 |
+
"grad_norm": 2.142010450363159,
|
| 826 |
+
"learning_rate": 0.00012704349399770767,
|
| 827 |
+
"loss": 8.3065,
|
| 828 |
+
"step": 2340
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.04271106687177631,
|
| 832 |
+
"grad_norm": 2.1896350383758545,
|
| 833 |
+
"learning_rate": 0.0001281293358267479,
|
| 834 |
+
"loss": 8.2688,
|
| 835 |
+
"step": 2360
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.043073025065604925,
|
| 839 |
+
"grad_norm": 2.1078433990478516,
|
| 840 |
+
"learning_rate": 0.00012921517765578815,
|
| 841 |
+
"loss": 8.2639,
|
| 842 |
+
"step": 2380
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.043434983259433534,
|
| 846 |
+
"grad_norm": 1.8464548587799072,
|
| 847 |
+
"learning_rate": 0.0001303010194848284,
|
| 848 |
+
"loss": 8.2638,
|
| 849 |
+
"step": 2400
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.04379694145326215,
|
| 853 |
+
"grad_norm": 2.71945858001709,
|
| 854 |
+
"learning_rate": 0.0001313868613138686,
|
| 855 |
+
"loss": 8.2516,
|
| 856 |
+
"step": 2420
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.044158899647090764,
|
| 860 |
+
"grad_norm": 1.1496859788894653,
|
| 861 |
+
"learning_rate": 0.00013247270314290885,
|
| 862 |
+
"loss": 8.2561,
|
| 863 |
+
"step": 2440
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.04452085784091937,
|
| 867 |
+
"grad_norm": 2.2801716327667236,
|
| 868 |
+
"learning_rate": 0.0001335585449719491,
|
| 869 |
+
"loss": 8.2525,
|
| 870 |
+
"step": 2460
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.04488281603474799,
|
| 874 |
+
"grad_norm": 2.1865906715393066,
|
| 875 |
+
"learning_rate": 0.00013464438680098933,
|
| 876 |
+
"loss": 8.2548,
|
| 877 |
+
"step": 2480
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.0452447742285766,
|
| 881 |
+
"grad_norm": 1.8173776865005493,
|
| 882 |
+
"learning_rate": 0.00013573022863002957,
|
| 883 |
+
"loss": 8.2447,
|
| 884 |
+
"step": 2500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.04560673242240521,
|
| 888 |
+
"grad_norm": 2.018167018890381,
|
| 889 |
+
"learning_rate": 0.0001368160704590698,
|
| 890 |
+
"loss": 8.2116,
|
| 891 |
+
"step": 2520
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.04596869061623383,
|
| 895 |
+
"grad_norm": 2.387749433517456,
|
| 896 |
+
"learning_rate": 0.00013790191228811003,
|
| 897 |
+
"loss": 8.2071,
|
| 898 |
+
"step": 2540
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.046330648810062436,
|
| 902 |
+
"grad_norm": 2.1164238452911377,
|
| 903 |
+
"learning_rate": 0.00013898775411715027,
|
| 904 |
+
"loss": 8.2173,
|
| 905 |
+
"step": 2560
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.04669260700389105,
|
| 909 |
+
"grad_norm": 2.6271204948425293,
|
| 910 |
+
"learning_rate": 0.0001400735959461905,
|
| 911 |
+
"loss": 8.1928,
|
| 912 |
+
"step": 2580
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.04705456519771967,
|
| 916 |
+
"grad_norm": 2.146430730819702,
|
| 917 |
+
"learning_rate": 0.00014115943777523075,
|
| 918 |
+
"loss": 8.2093,
|
| 919 |
+
"step": 2600
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.047416523391548275,
|
| 923 |
+
"grad_norm": 1.758144736289978,
|
| 924 |
+
"learning_rate": 0.000142245279604271,
|
| 925 |
+
"loss": 8.1709,
|
| 926 |
+
"step": 2620
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.04777848158537689,
|
| 930 |
+
"grad_norm": 1.3466659784317017,
|
| 931 |
+
"learning_rate": 0.00014333112143331123,
|
| 932 |
+
"loss": 8.1766,
|
| 933 |
+
"step": 2640
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.0481404397792055,
|
| 937 |
+
"grad_norm": 1.9450665712356567,
|
| 938 |
+
"learning_rate": 0.00014441696326235144,
|
| 939 |
+
"loss": 8.1901,
|
| 940 |
+
"step": 2660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.048502397973034114,
|
| 944 |
+
"grad_norm": 1.6330885887145996,
|
| 945 |
+
"learning_rate": 0.00014550280509139168,
|
| 946 |
+
"loss": 8.1911,
|
| 947 |
+
"step": 2680
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.04886435616686273,
|
| 951 |
+
"grad_norm": 1.8187795877456665,
|
| 952 |
+
"learning_rate": 0.00014658864692043192,
|
| 953 |
+
"loss": 8.1737,
|
| 954 |
+
"step": 2700
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.04922631436069134,
|
| 958 |
+
"grad_norm": 2.8557980060577393,
|
| 959 |
+
"learning_rate": 0.00014767448874947216,
|
| 960 |
+
"loss": 8.1732,
|
| 961 |
+
"step": 2720
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.049588272554519953,
|
| 965 |
+
"grad_norm": 2.0480148792266846,
|
| 966 |
+
"learning_rate": 0.0001487603305785124,
|
| 967 |
+
"loss": 8.1636,
|
| 968 |
+
"step": 2740
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.04995023074834857,
|
| 972 |
+
"grad_norm": 1.8413054943084717,
|
| 973 |
+
"learning_rate": 0.00014984617240755265,
|
| 974 |
+
"loss": 8.1644,
|
| 975 |
+
"step": 2760
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.05031218894217718,
|
| 979 |
+
"grad_norm": 1.5977145433425903,
|
| 980 |
+
"learning_rate": 0.00015093201423659289,
|
| 981 |
+
"loss": 8.1652,
|
| 982 |
+
"step": 2780
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.05067414713600579,
|
| 986 |
+
"grad_norm": 2.060908317565918,
|
| 987 |
+
"learning_rate": 0.0001520178560656331,
|
| 988 |
+
"loss": 8.1747,
|
| 989 |
+
"step": 2800
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.0510361053298344,
|
| 993 |
+
"grad_norm": 2.097968339920044,
|
| 994 |
+
"learning_rate": 0.00015310369789467334,
|
| 995 |
+
"loss": 8.1595,
|
| 996 |
+
"step": 2820
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.051398063523663017,
|
| 1000 |
+
"grad_norm": 2.275170087814331,
|
| 1001 |
+
"learning_rate": 0.00015418953972371358,
|
| 1002 |
+
"loss": 8.1677,
|
| 1003 |
+
"step": 2840
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.05176002171749163,
|
| 1007 |
+
"grad_norm": 1.372065544128418,
|
| 1008 |
+
"learning_rate": 0.00015527538155275382,
|
| 1009 |
+
"loss": 8.14,
|
| 1010 |
+
"step": 2860
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.05212197991132024,
|
| 1014 |
+
"grad_norm": 1.472987174987793,
|
| 1015 |
+
"learning_rate": 0.00015636122338179406,
|
| 1016 |
+
"loss": 8.1664,
|
| 1017 |
+
"step": 2880
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.052483938105148856,
|
| 1021 |
+
"grad_norm": 1.398430347442627,
|
| 1022 |
+
"learning_rate": 0.0001574470652108343,
|
| 1023 |
+
"loss": 8.1435,
|
| 1024 |
+
"step": 2900
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.05284589629897747,
|
| 1028 |
+
"grad_norm": 2.2276878356933594,
|
| 1029 |
+
"learning_rate": 0.00015853290703987452,
|
| 1030 |
+
"loss": 8.1588,
|
| 1031 |
+
"step": 2920
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.05320785449280608,
|
| 1035 |
+
"grad_norm": 2.8768556118011475,
|
| 1036 |
+
"learning_rate": 0.00015961874886891476,
|
| 1037 |
+
"loss": 8.15,
|
| 1038 |
+
"step": 2940
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.053569812686634695,
|
| 1042 |
+
"grad_norm": 2.1943325996398926,
|
| 1043 |
+
"learning_rate": 0.000160704590697955,
|
| 1044 |
+
"loss": 8.1515,
|
| 1045 |
+
"step": 2960
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.0539317708804633,
|
| 1049 |
+
"grad_norm": 2.0583720207214355,
|
| 1050 |
+
"learning_rate": 0.00016179043252699524,
|
| 1051 |
+
"loss": 8.1343,
|
| 1052 |
+
"step": 2980
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.05429372907429192,
|
| 1056 |
+
"grad_norm": 2.024512767791748,
|
| 1057 |
+
"learning_rate": 0.00016282198226458345,
|
| 1058 |
+
"loss": 8.1324,
|
| 1059 |
+
"step": 3000
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.054655687268120534,
|
| 1063 |
+
"grad_norm": 2.0019822120666504,
|
| 1064 |
+
"learning_rate": 0.0001639078240936237,
|
| 1065 |
+
"loss": 8.1268,
|
| 1066 |
+
"step": 3020
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.05501764546194914,
|
| 1070 |
+
"grad_norm": 2.0442886352539062,
|
| 1071 |
+
"learning_rate": 0.00016499366592266393,
|
| 1072 |
+
"loss": 8.1173,
|
| 1073 |
+
"step": 3040
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.05537960365577776,
|
| 1077 |
+
"grad_norm": 2.780470371246338,
|
| 1078 |
+
"learning_rate": 0.00016607950775170417,
|
| 1079 |
+
"loss": 8.1345,
|
| 1080 |
+
"step": 3060
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.05574156184960637,
|
| 1084 |
+
"grad_norm": 2.3489673137664795,
|
| 1085 |
+
"learning_rate": 0.00016716534958074441,
|
| 1086 |
+
"loss": 8.1356,
|
| 1087 |
+
"step": 3080
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.05610352004343498,
|
| 1091 |
+
"grad_norm": 1.5495760440826416,
|
| 1092 |
+
"learning_rate": 0.00016825119140978465,
|
| 1093 |
+
"loss": 8.1344,
|
| 1094 |
+
"step": 3100
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.0564654782372636,
|
| 1098 |
+
"grad_norm": 1.9256787300109863,
|
| 1099 |
+
"learning_rate": 0.00016933703323882487,
|
| 1100 |
+
"loss": 8.1298,
|
| 1101 |
+
"step": 3120
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.056827436431092206,
|
| 1105 |
+
"grad_norm": 2.0177950859069824,
|
| 1106 |
+
"learning_rate": 0.0001704228750678651,
|
| 1107 |
+
"loss": 8.1168,
|
| 1108 |
+
"step": 3140
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.05718939462492082,
|
| 1112 |
+
"grad_norm": 2.141857147216797,
|
| 1113 |
+
"learning_rate": 0.00017150871689690535,
|
| 1114 |
+
"loss": 8.1286,
|
| 1115 |
+
"step": 3160
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.057551352818749436,
|
| 1119 |
+
"grad_norm": 2.1056764125823975,
|
| 1120 |
+
"learning_rate": 0.0001725945587259456,
|
| 1121 |
+
"loss": 8.1398,
|
| 1122 |
+
"step": 3180
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.057913311012578045,
|
| 1126 |
+
"grad_norm": 1.9108024835586548,
|
| 1127 |
+
"learning_rate": 0.00017368040055498583,
|
| 1128 |
+
"loss": 8.1244,
|
| 1129 |
+
"step": 3200
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.05827526920640666,
|
| 1133 |
+
"grad_norm": 1.7270424365997314,
|
| 1134 |
+
"learning_rate": 0.00017476624238402607,
|
| 1135 |
+
"loss": 8.1102,
|
| 1136 |
+
"step": 3220
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.058637227400235276,
|
| 1140 |
+
"grad_norm": 1.684164047241211,
|
| 1141 |
+
"learning_rate": 0.00017585208421306629,
|
| 1142 |
+
"loss": 8.1079,
|
| 1143 |
+
"step": 3240
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.058999185594063884,
|
| 1147 |
+
"grad_norm": 2.553480625152588,
|
| 1148 |
+
"learning_rate": 0.00017688363395065452,
|
| 1149 |
+
"loss": 8.1094,
|
| 1150 |
+
"step": 3260
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.0593611437878925,
|
| 1154 |
+
"grad_norm": 1.9391483068466187,
|
| 1155 |
+
"learning_rate": 0.00017796947577969477,
|
| 1156 |
+
"loss": 8.1078,
|
| 1157 |
+
"step": 3280
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.05972310198172111,
|
| 1161 |
+
"grad_norm": 2.539398193359375,
|
| 1162 |
+
"learning_rate": 0.000179055317608735,
|
| 1163 |
+
"loss": 8.1159,
|
| 1164 |
+
"step": 3300
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.06008506017554972,
|
| 1168 |
+
"grad_norm": 2.3800413608551025,
|
| 1169 |
+
"learning_rate": 0.00018014115943777525,
|
| 1170 |
+
"loss": 8.1043,
|
| 1171 |
+
"step": 3320
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.06044701836937834,
|
| 1175 |
+
"grad_norm": 2.1216628551483154,
|
| 1176 |
+
"learning_rate": 0.00018122700126681546,
|
| 1177 |
+
"loss": 8.0919,
|
| 1178 |
+
"step": 3340
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.06080897656320695,
|
| 1182 |
+
"grad_norm": 3.430650472640991,
|
| 1183 |
+
"learning_rate": 0.0001823128430958557,
|
| 1184 |
+
"loss": 8.1078,
|
| 1185 |
+
"step": 3360
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.06117093475703556,
|
| 1189 |
+
"grad_norm": 2.1858770847320557,
|
| 1190 |
+
"learning_rate": 0.00018339868492489594,
|
| 1191 |
+
"loss": 8.1184,
|
| 1192 |
+
"step": 3380
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.06153289295086418,
|
| 1196 |
+
"grad_norm": 2.896089553833008,
|
| 1197 |
+
"learning_rate": 0.00018448452675393618,
|
| 1198 |
+
"loss": 8.106,
|
| 1199 |
+
"step": 3400
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.061894851144692786,
|
| 1203 |
+
"grad_norm": 2.7624003887176514,
|
| 1204 |
+
"learning_rate": 0.00018557036858297642,
|
| 1205 |
+
"loss": 8.1072,
|
| 1206 |
+
"step": 3420
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.0622568093385214,
|
| 1210 |
+
"grad_norm": 2.464115619659424,
|
| 1211 |
+
"learning_rate": 0.00018665621041201666,
|
| 1212 |
+
"loss": 8.0962,
|
| 1213 |
+
"step": 3440
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.06261876753235002,
|
| 1217 |
+
"grad_norm": 2.5943491458892822,
|
| 1218 |
+
"learning_rate": 0.00018774205224105688,
|
| 1219 |
+
"loss": 8.0914,
|
| 1220 |
+
"step": 3460
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.06298072572617863,
|
| 1224 |
+
"grad_norm": 2.0824356079101562,
|
| 1225 |
+
"learning_rate": 0.00018882789407009712,
|
| 1226 |
+
"loss": 8.1222,
|
| 1227 |
+
"step": 3480
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.06334268392000723,
|
| 1231 |
+
"grad_norm": 2.8781402111053467,
|
| 1232 |
+
"learning_rate": 0.00018991373589913736,
|
| 1233 |
+
"loss": 8.088,
|
| 1234 |
+
"step": 3500
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.06370464211383585,
|
| 1238 |
+
"grad_norm": 2.0000219345092773,
|
| 1239 |
+
"learning_rate": 0.0001909995777281776,
|
| 1240 |
+
"loss": 8.103,
|
| 1241 |
+
"step": 3520
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.06406660030766446,
|
| 1245 |
+
"grad_norm": 2.4691524505615234,
|
| 1246 |
+
"learning_rate": 0.00019208541955721784,
|
| 1247 |
+
"loss": 8.1045,
|
| 1248 |
+
"step": 3540
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.06442855850149308,
|
| 1252 |
+
"grad_norm": 2.583723545074463,
|
| 1253 |
+
"learning_rate": 0.00019317126138625808,
|
| 1254 |
+
"loss": 8.0879,
|
| 1255 |
+
"step": 3560
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.0647905166953217,
|
| 1259 |
+
"grad_norm": 2.7288269996643066,
|
| 1260 |
+
"learning_rate": 0.0001942571032152983,
|
| 1261 |
+
"loss": 8.091,
|
| 1262 |
+
"step": 3580
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.0651524748891503,
|
| 1266 |
+
"grad_norm": 2.360276699066162,
|
| 1267 |
+
"learning_rate": 0.00019534294504433854,
|
| 1268 |
+
"loss": 8.0894,
|
| 1269 |
+
"step": 3600
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.06551443308297891,
|
| 1273 |
+
"grad_norm": 2.6591217517852783,
|
| 1274 |
+
"learning_rate": 0.00019642878687337878,
|
| 1275 |
+
"loss": 8.082,
|
| 1276 |
+
"step": 3620
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.06587639127680753,
|
| 1280 |
+
"grad_norm": 2.5572097301483154,
|
| 1281 |
+
"learning_rate": 0.00019751462870241902,
|
| 1282 |
+
"loss": 8.1033,
|
| 1283 |
+
"step": 3640
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.06623834947063614,
|
| 1287 |
+
"grad_norm": 2.643139600753784,
|
| 1288 |
+
"learning_rate": 0.00019860047053145926,
|
| 1289 |
+
"loss": 8.1124,
|
| 1290 |
+
"step": 3660
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.06660030766446476,
|
| 1294 |
+
"grad_norm": 3.0322484970092773,
|
| 1295 |
+
"learning_rate": 0.0001996863123604995,
|
| 1296 |
+
"loss": 8.1049,
|
| 1297 |
+
"step": 3680
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.06696226585829337,
|
| 1301 |
+
"grad_norm": 2.7740566730499268,
|
| 1302 |
+
"learning_rate": 0.0002007721541895397,
|
| 1303 |
+
"loss": 8.0876,
|
| 1304 |
+
"step": 3700
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.06732422405212198,
|
| 1308 |
+
"grad_norm": 2.7816436290740967,
|
| 1309 |
+
"learning_rate": 0.00020185799601857995,
|
| 1310 |
+
"loss": 8.1121,
|
| 1311 |
+
"step": 3720
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.06768618224595059,
|
| 1315 |
+
"grad_norm": 2.4198217391967773,
|
| 1316 |
+
"learning_rate": 0.0002029438378476202,
|
| 1317 |
+
"loss": 8.0958,
|
| 1318 |
+
"step": 3740
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.0680481404397792,
|
| 1322 |
+
"grad_norm": 2.8162808418273926,
|
| 1323 |
+
"learning_rate": 0.00020402967967666043,
|
| 1324 |
+
"loss": 8.0897,
|
| 1325 |
+
"step": 3760
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.06841009863360782,
|
| 1329 |
+
"grad_norm": 2.883631706237793,
|
| 1330 |
+
"learning_rate": 0.00020511552150570067,
|
| 1331 |
+
"loss": 8.085,
|
| 1332 |
+
"step": 3780
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.06877205682743644,
|
| 1336 |
+
"grad_norm": 2.4725236892700195,
|
| 1337 |
+
"learning_rate": 0.00020620136333474092,
|
| 1338 |
+
"loss": 8.0807,
|
| 1339 |
+
"step": 3800
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.06913401502126504,
|
| 1343 |
+
"grad_norm": 2.3156867027282715,
|
| 1344 |
+
"learning_rate": 0.00020728720516378116,
|
| 1345 |
+
"loss": 8.0747,
|
| 1346 |
+
"step": 3820
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.06949597321509365,
|
| 1350 |
+
"grad_norm": 3.308699607849121,
|
| 1351 |
+
"learning_rate": 0.00020837304699282137,
|
| 1352 |
+
"loss": 8.0767,
|
| 1353 |
+
"step": 3840
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.06985793140892227,
|
| 1357 |
+
"grad_norm": 3.143287181854248,
|
| 1358 |
+
"learning_rate": 0.0002094588888218616,
|
| 1359 |
+
"loss": 8.0831,
|
| 1360 |
+
"step": 3860
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.07021988960275088,
|
| 1364 |
+
"grad_norm": 3.100562810897827,
|
| 1365 |
+
"learning_rate": 0.00021054473065090185,
|
| 1366 |
+
"loss": 8.0862,
|
| 1367 |
+
"step": 3880
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.0705818477965795,
|
| 1371 |
+
"grad_norm": 2.48494029045105,
|
| 1372 |
+
"learning_rate": 0.0002116305724799421,
|
| 1373 |
+
"loss": 8.0952,
|
| 1374 |
+
"step": 3900
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.0709438059904081,
|
| 1378 |
+
"grad_norm": 3.1432759761810303,
|
| 1379 |
+
"learning_rate": 0.00021271641430898233,
|
| 1380 |
+
"loss": 8.0906,
|
| 1381 |
+
"step": 3920
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.07130576418423672,
|
| 1385 |
+
"grad_norm": 3.330761194229126,
|
| 1386 |
+
"learning_rate": 0.00021380225613802257,
|
| 1387 |
+
"loss": 8.0856,
|
| 1388 |
+
"step": 3940
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.07166772237806533,
|
| 1392 |
+
"grad_norm": 3.6338093280792236,
|
| 1393 |
+
"learning_rate": 0.00021488809796706279,
|
| 1394 |
+
"loss": 8.0733,
|
| 1395 |
+
"step": 3960
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.07202968057189395,
|
| 1399 |
+
"grad_norm": 3.014366388320923,
|
| 1400 |
+
"learning_rate": 0.00021597393979610303,
|
| 1401 |
+
"loss": 8.086,
|
| 1402 |
+
"step": 3980
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.07239163876572256,
|
| 1406 |
+
"grad_norm": 2.774247169494629,
|
| 1407 |
+
"learning_rate": 0.00021705978162514327,
|
| 1408 |
+
"loss": 8.093,
|
| 1409 |
+
"step": 4000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.07275359695955118,
|
| 1413 |
+
"grad_norm": 4.0621867179870605,
|
| 1414 |
+
"learning_rate": 0.0002181456234541835,
|
| 1415 |
+
"loss": 8.0801,
|
| 1416 |
+
"step": 4020
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.07311555515337978,
|
| 1420 |
+
"grad_norm": 3.9556710720062256,
|
| 1421 |
+
"learning_rate": 0.00021923146528322375,
|
| 1422 |
+
"loss": 8.0632,
|
| 1423 |
+
"step": 4040
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.0734775133472084,
|
| 1427 |
+
"grad_norm": 3.0062179565429688,
|
| 1428 |
+
"learning_rate": 0.000220317307112264,
|
| 1429 |
+
"loss": 8.0713,
|
| 1430 |
+
"step": 4060
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.07383947154103701,
|
| 1434 |
+
"grad_norm": 3.4333982467651367,
|
| 1435 |
+
"learning_rate": 0.0002214031489413042,
|
| 1436 |
+
"loss": 8.0656,
|
| 1437 |
+
"step": 4080
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.07420142973486563,
|
| 1441 |
+
"grad_norm": 3.107091188430786,
|
| 1442 |
+
"learning_rate": 0.00022248899077034444,
|
| 1443 |
+
"loss": 8.0832,
|
| 1444 |
+
"step": 4100
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.07456338792869424,
|
| 1448 |
+
"grad_norm": 3.5279381275177,
|
| 1449 |
+
"learning_rate": 0.00022357483259938468,
|
| 1450 |
+
"loss": 8.0703,
|
| 1451 |
+
"step": 4120
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.07492534612252284,
|
| 1455 |
+
"grad_norm": 2.9503841400146484,
|
| 1456 |
+
"learning_rate": 0.00022466067442842493,
|
| 1457 |
+
"loss": 8.0796,
|
| 1458 |
+
"step": 4140
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.07528730431635146,
|
| 1462 |
+
"grad_norm": 3.018066644668579,
|
| 1463 |
+
"learning_rate": 0.00022574651625746517,
|
| 1464 |
+
"loss": 8.07,
|
| 1465 |
+
"step": 4160
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.07564926251018007,
|
| 1469 |
+
"grad_norm": 3.552546501159668,
|
| 1470 |
+
"learning_rate": 0.0002268323580865054,
|
| 1471 |
+
"loss": 8.1046,
|
| 1472 |
+
"step": 4180
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.07601122070400869,
|
| 1476 |
+
"grad_norm": 3.881967306137085,
|
| 1477 |
+
"learning_rate": 0.00022791819991554565,
|
| 1478 |
+
"loss": 8.0622,
|
| 1479 |
+
"step": 4200
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.0763731788978373,
|
| 1483 |
+
"grad_norm": 4.438878536224365,
|
| 1484 |
+
"learning_rate": 0.0002290040417445859,
|
| 1485 |
+
"loss": 8.0783,
|
| 1486 |
+
"step": 4220
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.0767351370916659,
|
| 1490 |
+
"grad_norm": 3.928950071334839,
|
| 1491 |
+
"learning_rate": 0.00023008988357362613,
|
| 1492 |
+
"loss": 8.0887,
|
| 1493 |
+
"step": 4240
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.07709709528549452,
|
| 1497 |
+
"grad_norm": 4.781463623046875,
|
| 1498 |
+
"learning_rate": 0.00023117572540266632,
|
| 1499 |
+
"loss": 8.0672,
|
| 1500 |
+
"step": 4260
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.07745905347932314,
|
| 1504 |
+
"grad_norm": 3.8346338272094727,
|
| 1505 |
+
"learning_rate": 0.00023226156723170656,
|
| 1506 |
+
"loss": 8.0822,
|
| 1507 |
+
"step": 4280
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.07782101167315175,
|
| 1511 |
+
"grad_norm": 3.835999011993408,
|
| 1512 |
+
"learning_rate": 0.0002333474090607468,
|
| 1513 |
+
"loss": 8.0662,
|
| 1514 |
+
"step": 4300
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.07818296986698037,
|
| 1518 |
+
"grad_norm": 4.432645320892334,
|
| 1519 |
+
"learning_rate": 0.00023443325088978704,
|
| 1520 |
+
"loss": 8.0816,
|
| 1521 |
+
"step": 4320
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.07854492806080898,
|
| 1525 |
+
"grad_norm": 3.856933116912842,
|
| 1526 |
+
"learning_rate": 0.00023551909271882728,
|
| 1527 |
+
"loss": 8.055,
|
| 1528 |
+
"step": 4340
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.07890688625463758,
|
| 1532 |
+
"grad_norm": 4.055251598358154,
|
| 1533 |
+
"learning_rate": 0.00023660493454786752,
|
| 1534 |
+
"loss": 8.0833,
|
| 1535 |
+
"step": 4360
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.0792688444484662,
|
| 1539 |
+
"grad_norm": 4.129009246826172,
|
| 1540 |
+
"learning_rate": 0.00023769077637690776,
|
| 1541 |
+
"loss": 8.0715,
|
| 1542 |
+
"step": 4380
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.07963080264229482,
|
| 1546 |
+
"grad_norm": 3.944307565689087,
|
| 1547 |
+
"learning_rate": 0.000238776618205948,
|
| 1548 |
+
"loss": 8.0721,
|
| 1549 |
+
"step": 4400
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.07999276083612343,
|
| 1553 |
+
"grad_norm": 4.226454257965088,
|
| 1554 |
+
"learning_rate": 0.00023986246003498824,
|
| 1555 |
+
"loss": 8.081,
|
| 1556 |
+
"step": 4420
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.08035471902995205,
|
| 1560 |
+
"grad_norm": 4.180859088897705,
|
| 1561 |
+
"learning_rate": 0.00024094830186402848,
|
| 1562 |
+
"loss": 8.0674,
|
| 1563 |
+
"step": 4440
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.08071667722378065,
|
| 1567 |
+
"grad_norm": 4.678595066070557,
|
| 1568 |
+
"learning_rate": 0.0002418712674187127,
|
| 1569 |
+
"loss": 8.6491,
|
| 1570 |
+
"step": 4460
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.08107863541760926,
|
| 1574 |
+
"grad_norm": 3.916884422302246,
|
| 1575 |
+
"learning_rate": 0.00024295710924775293,
|
| 1576 |
+
"loss": 8.0667,
|
| 1577 |
+
"step": 4480
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.08144059361143788,
|
| 1581 |
+
"grad_norm": 5.014923572540283,
|
| 1582 |
+
"learning_rate": 0.00024404295107679317,
|
| 1583 |
+
"loss": 8.0709,
|
| 1584 |
+
"step": 4500
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.0818025518052665,
|
| 1588 |
+
"grad_norm": 5.424384593963623,
|
| 1589 |
+
"learning_rate": 0.00024512879290583333,
|
| 1590 |
+
"loss": 8.0714,
|
| 1591 |
+
"step": 4520
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.08216450999909511,
|
| 1595 |
+
"grad_norm": 4.7506890296936035,
|
| 1596 |
+
"learning_rate": 0.0002462146347348736,
|
| 1597 |
+
"loss": 8.1164,
|
| 1598 |
+
"step": 4540
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.08252646819292371,
|
| 1602 |
+
"grad_norm": 4.271077632904053,
|
| 1603 |
+
"learning_rate": 0.0002473004765639138,
|
| 1604 |
+
"loss": 8.0456,
|
| 1605 |
+
"step": 4560
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.08288842638675233,
|
| 1609 |
+
"grad_norm": 5.108422756195068,
|
| 1610 |
+
"learning_rate": 0.00024838631839295405,
|
| 1611 |
+
"loss": 8.0683,
|
| 1612 |
+
"step": 4580
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.08325038458058094,
|
| 1616 |
+
"grad_norm": 4.00137186050415,
|
| 1617 |
+
"learning_rate": 0.0002494721602219943,
|
| 1618 |
+
"loss": 8.0877,
|
| 1619 |
+
"step": 4600
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.08361234277440956,
|
| 1623 |
+
"grad_norm": 5.020883560180664,
|
| 1624 |
+
"learning_rate": 0.00025055800205103454,
|
| 1625 |
+
"loss": 8.0669,
|
| 1626 |
+
"step": 4620
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.08397430096823817,
|
| 1630 |
+
"grad_norm": 4.774672985076904,
|
| 1631 |
+
"learning_rate": 0.0002516438438800748,
|
| 1632 |
+
"loss": 8.0978,
|
| 1633 |
+
"step": 4640
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.08433625916206679,
|
| 1637 |
+
"grad_norm": 4.414867877960205,
|
| 1638 |
+
"learning_rate": 0.000252729685709115,
|
| 1639 |
+
"loss": 8.0909,
|
| 1640 |
+
"step": 4660
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.08469821735589539,
|
| 1644 |
+
"grad_norm": 4.128323554992676,
|
| 1645 |
+
"learning_rate": 0.00025381552753815526,
|
| 1646 |
+
"loss": 8.09,
|
| 1647 |
+
"step": 4680
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.085060175549724,
|
| 1651 |
+
"grad_norm": 4.486546039581299,
|
| 1652 |
+
"learning_rate": 0.0002549013693671955,
|
| 1653 |
+
"loss": 8.0744,
|
| 1654 |
+
"step": 4700
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.08542213374355262,
|
| 1658 |
+
"grad_norm": 4.218783855438232,
|
| 1659 |
+
"learning_rate": 0.00025598721119623574,
|
| 1660 |
+
"loss": 8.0902,
|
| 1661 |
+
"step": 4720
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.08578409193738123,
|
| 1665 |
+
"grad_norm": 4.9139084815979,
|
| 1666 |
+
"learning_rate": 0.000257073053025276,
|
| 1667 |
+
"loss": 8.0726,
|
| 1668 |
+
"step": 4740
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.08614605013120985,
|
| 1672 |
+
"grad_norm": 4.544093132019043,
|
| 1673 |
+
"learning_rate": 0.0002581588948543162,
|
| 1674 |
+
"loss": 8.0735,
|
| 1675 |
+
"step": 4760
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.08650800832503845,
|
| 1679 |
+
"grad_norm": 5.45837926864624,
|
| 1680 |
+
"learning_rate": 0.0002592447366833564,
|
| 1681 |
+
"loss": 8.0781,
|
| 1682 |
+
"step": 4780
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.08686996651886707,
|
| 1686 |
+
"grad_norm": 5.2784423828125,
|
| 1687 |
+
"learning_rate": 0.00026033057851239665,
|
| 1688 |
+
"loss": 8.088,
|
| 1689 |
+
"step": 4800
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.08723192471269568,
|
| 1693 |
+
"grad_norm": 4.507415294647217,
|
| 1694 |
+
"learning_rate": 0.0002614164203414369,
|
| 1695 |
+
"loss": 8.0707,
|
| 1696 |
+
"step": 4820
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.0875938829065243,
|
| 1700 |
+
"grad_norm": 4.857511520385742,
|
| 1701 |
+
"learning_rate": 0.00026250226217047713,
|
| 1702 |
+
"loss": 8.0994,
|
| 1703 |
+
"step": 4840
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.08795584110035291,
|
| 1707 |
+
"grad_norm": 4.420199871063232,
|
| 1708 |
+
"learning_rate": 0.00026358810399951737,
|
| 1709 |
+
"loss": 8.0605,
|
| 1710 |
+
"step": 4860
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.08831779929418153,
|
| 1714 |
+
"grad_norm": 3.7216994762420654,
|
| 1715 |
+
"learning_rate": 0.0002646739458285576,
|
| 1716 |
+
"loss": 8.1182,
|
| 1717 |
+
"step": 4880
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.08867975748801013,
|
| 1721 |
+
"grad_norm": 4.462796688079834,
|
| 1722 |
+
"learning_rate": 0.00026575978765759785,
|
| 1723 |
+
"loss": 8.0666,
|
| 1724 |
+
"step": 4900
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.08904171568183875,
|
| 1728 |
+
"grad_norm": 4.29760217666626,
|
| 1729 |
+
"learning_rate": 0.0002668456294866381,
|
| 1730 |
+
"loss": 8.0548,
|
| 1731 |
+
"step": 4920
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.08940367387566736,
|
| 1735 |
+
"grad_norm": 5.2155046463012695,
|
| 1736 |
+
"learning_rate": 0.00026793147131567833,
|
| 1737 |
+
"loss": 8.1614,
|
| 1738 |
+
"step": 4940
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.08976563206949598,
|
| 1742 |
+
"grad_norm": 4.687706470489502,
|
| 1743 |
+
"learning_rate": 0.0002690173131447186,
|
| 1744 |
+
"loss": 8.051,
|
| 1745 |
+
"step": 4960
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.09012759026332459,
|
| 1749 |
+
"grad_norm": 3.9040534496307373,
|
| 1750 |
+
"learning_rate": 0.0002701031549737588,
|
| 1751 |
+
"loss": 8.0645,
|
| 1752 |
+
"step": 4980
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.0904895484571532,
|
| 1756 |
+
"grad_norm": 4.2080159187316895,
|
| 1757 |
+
"learning_rate": 0.00027118899680279905,
|
| 1758 |
+
"loss": 8.059,
|
| 1759 |
+
"step": 5000
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.0904895484571532,
|
| 1763 |
+
"eval_accuracy": 0.10885986383166214,
|
| 1764 |
+
"eval_loss": 8.154509544372559,
|
| 1765 |
+
"eval_runtime": 172.8056,
|
| 1766 |
+
"eval_samples_per_second": 3517.479,
|
| 1767 |
+
"eval_steps_per_second": 3.437,
|
| 1768 |
+
"step": 5000
|
| 1769 |
+
}
|
| 1770 |
+
],
|
| 1771 |
+
"logging_steps": 20,
|
| 1772 |
+
"max_steps": 165765,
|
| 1773 |
+
"num_input_tokens_seen": 0,
|
| 1774 |
+
"num_train_epochs": 3,
|
| 1775 |
+
"save_steps": 5000,
|
| 1776 |
+
"stateful_callbacks": {
|
| 1777 |
+
"TrainerControl": {
|
| 1778 |
+
"args": {
|
| 1779 |
+
"should_epoch_stop": false,
|
| 1780 |
+
"should_evaluate": false,
|
| 1781 |
+
"should_log": false,
|
| 1782 |
+
"should_save": false,
|
| 1783 |
+
"should_training_stop": false
|
| 1784 |
+
},
|
| 1785 |
+
"attributes": {}
|
| 1786 |
+
}
|
| 1787 |
+
},
|
| 1788 |
+
"total_flos": 7191810539520000.0,
|
| 1789 |
+
"train_batch_size": 512,
|
| 1790 |
+
"trial_name": null,
|
| 1791 |
+
"trial_params": null
|
| 1792 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f84e7b8139fc15504603acf198a1d0e2de337ab7857597ceb1e053e7d16b4de6
|
| 3 |
+
size 5176
|