Sepedi-Llama v1 - loss 83 to 19
Browse files- .gitattributes +3 -0
- checkpoint-11500/config.json +30 -0
- checkpoint-11500/model.safetensors +3 -0
- checkpoint-11500/optimizer.pt +3 -0
- checkpoint-11500/rng_state.pth +3 -0
- checkpoint-11500/scheduler.pt +3 -0
- checkpoint-11500/tokenizer.json +3 -0
- checkpoint-11500/tokenizer_config.json +15 -0
- checkpoint-11500/trainer_state.json +1644 -0
- checkpoint-11500/training_args.bin +3 -0
- checkpoint-11776/config.json +30 -0
- checkpoint-11776/model.safetensors +3 -0
- checkpoint-11776/optimizer.pt +3 -0
- checkpoint-11776/rng_state.pth +3 -0
- checkpoint-11776/scheduler.pt +3 -0
- checkpoint-11776/tokenizer.json +3 -0
- checkpoint-11776/tokenizer_config.json +15 -0
- checkpoint-11776/trainer_state.json +1679 -0
- checkpoint-11776/training_args.bin +3 -0
- config.json +30 -0
- model.safetensors +3 -0
- tokenizer.json +3 -0
- tokenizer_config.json +15 -0
- training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
checkpoint-11500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
checkpoint-11776/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
checkpoint-11500/config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_cross_attention": false,
|
| 3 |
+
"architectures": [
|
| 4 |
+
"XLMRobertaForMaskedLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"dtype": "float32",
|
| 10 |
+
"eos_token_id": 2,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.1,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"is_decoder": false,
|
| 17 |
+
"layer_norm_eps": 1e-05,
|
| 18 |
+
"max_position_embeddings": 514,
|
| 19 |
+
"model_type": "xlm-roberta",
|
| 20 |
+
"num_attention_heads": 12,
|
| 21 |
+
"num_hidden_layers": 12,
|
| 22 |
+
"output_past": true,
|
| 23 |
+
"pad_token_id": 1,
|
| 24 |
+
"position_embedding_type": "absolute",
|
| 25 |
+
"tie_word_embeddings": true,
|
| 26 |
+
"transformers_version": "5.10.2",
|
| 27 |
+
"type_vocab_size": 1,
|
| 28 |
+
"use_cache": false,
|
| 29 |
+
"vocab_size": 250002
|
| 30 |
+
}
|
checkpoint-11500/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d6e77f247066a23923a2da326e28c622b19d1db30f0c3a3a52d56564c4581452
|
| 3 |
+
size 1113205088
|
checkpoint-11500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cb0c07abf2730fe571049e8683d88bf124f11c51e399180ed59841d46674077d
|
| 3 |
+
size 2226533707
|
checkpoint-11500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8a244d1b2cabd27e6ef6833fab8f110fce186442b133311b8d8b7cf71e49c393
|
| 3 |
+
size 14645
|
checkpoint-11500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5aae1d3fe34313b912767627227deaade5ba5ba9a8c0205704029484c4296bd7
|
| 3 |
+
size 1465
|
checkpoint-11500/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:acbd420e2269cdc1ef45332d3d5c418be4aef6b8cb5a0b7ccae0893485307153
|
| 3 |
+
size 17098086
|
checkpoint-11500/tokenizer_config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": true,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<s>",
|
| 5 |
+
"cls_token": "<s>",
|
| 6 |
+
"eos_token": "</s>",
|
| 7 |
+
"is_local": false,
|
| 8 |
+
"local_files_only": false,
|
| 9 |
+
"mask_token": "<mask>",
|
| 10 |
+
"model_max_length": 512,
|
| 11 |
+
"pad_token": "<pad>",
|
| 12 |
+
"sep_token": "</s>",
|
| 13 |
+
"tokenizer_class": "XLMRobertaTokenizer",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
checkpoint-11500/trainer_state.json
ADDED
|
@@ -0,0 +1,1644 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.9766350675682849,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 11500,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.004246239424209934,
|
| 14 |
+
"grad_norm": 324.3861999511719,
|
| 15 |
+
"learning_rate": 1.225e-05,
|
| 16 |
+
"loss": 83.27494140625,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.008492478848419869,
|
| 21 |
+
"grad_norm": 294.6669921875,
|
| 22 |
+
"learning_rate": 2.4750000000000002e-05,
|
| 23 |
+
"loss": 72.928427734375,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.012738718272629802,
|
| 28 |
+
"grad_norm": 254.57310485839844,
|
| 29 |
+
"learning_rate": 3.7250000000000004e-05,
|
| 30 |
+
"loss": 65.434755859375,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.016984957696839737,
|
| 35 |
+
"grad_norm": 285.2427062988281,
|
| 36 |
+
"learning_rate": 4.975e-05,
|
| 37 |
+
"loss": 59.52314453125,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.02123119712104967,
|
| 42 |
+
"grad_norm": 276.11651611328125,
|
| 43 |
+
"learning_rate": 4.978835521769178e-05,
|
| 44 |
+
"loss": 56.26099609375,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.025477436545259604,
|
| 49 |
+
"grad_norm": 264.0826416015625,
|
| 50 |
+
"learning_rate": 4.9572391154111955e-05,
|
| 51 |
+
"loss": 52.364111328125,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.029723675969469537,
|
| 56 |
+
"grad_norm": 237.80140686035156,
|
| 57 |
+
"learning_rate": 4.9356427090532136e-05,
|
| 58 |
+
"loss": 49.438955078125,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.033969915393679474,
|
| 63 |
+
"grad_norm": 245.1451873779297,
|
| 64 |
+
"learning_rate": 4.914046302695232e-05,
|
| 65 |
+
"loss": 48.1695849609375,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.038216154817889404,
|
| 70 |
+
"grad_norm": 207.61068725585938,
|
| 71 |
+
"learning_rate": 4.892449896337249e-05,
|
| 72 |
+
"loss": 45.5499365234375,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.04246239424209934,
|
| 77 |
+
"grad_norm": 200.3720703125,
|
| 78 |
+
"learning_rate": 4.870853489979268e-05,
|
| 79 |
+
"loss": 44.5572998046875,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.04670863366630928,
|
| 84 |
+
"grad_norm": 197.68887329101562,
|
| 85 |
+
"learning_rate": 4.8492570836212855e-05,
|
| 86 |
+
"loss": 42.489599609375,
|
| 87 |
+
"step": 550
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.05095487309051921,
|
| 91 |
+
"grad_norm": 179.29957580566406,
|
| 92 |
+
"learning_rate": 4.827660677263304e-05,
|
| 93 |
+
"loss": 41.3450439453125,
|
| 94 |
+
"step": 600
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.055201112514729145,
|
| 98 |
+
"grad_norm": 226.55911254882812,
|
| 99 |
+
"learning_rate": 4.806064270905322e-05,
|
| 100 |
+
"loss": 41.2511376953125,
|
| 101 |
+
"step": 650
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.059447351938939075,
|
| 105 |
+
"grad_norm": 202.10791015625,
|
| 106 |
+
"learning_rate": 4.784467864547339e-05,
|
| 107 |
+
"loss": 39.70830322265625,
|
| 108 |
+
"step": 700
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.063693591363149,
|
| 112 |
+
"grad_norm": 207.3136444091797,
|
| 113 |
+
"learning_rate": 4.7628714581893575e-05,
|
| 114 |
+
"loss": 39.89173828125,
|
| 115 |
+
"step": 750
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.06793983078735895,
|
| 119 |
+
"grad_norm": 189.1178436279297,
|
| 120 |
+
"learning_rate": 4.7412750518313756e-05,
|
| 121 |
+
"loss": 38.6139501953125,
|
| 122 |
+
"step": 800
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.07218607021156888,
|
| 126 |
+
"grad_norm": 190.94015502929688,
|
| 127 |
+
"learning_rate": 4.719678645473393e-05,
|
| 128 |
+
"loss": 38.68489501953125,
|
| 129 |
+
"step": 850
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.07643230963577881,
|
| 133 |
+
"grad_norm": 191.75172424316406,
|
| 134 |
+
"learning_rate": 4.698082239115412e-05,
|
| 135 |
+
"loss": 36.644111328125,
|
| 136 |
+
"step": 900
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.08067854905998875,
|
| 140 |
+
"grad_norm": 206.9021453857422,
|
| 141 |
+
"learning_rate": 4.6764858327574294e-05,
|
| 142 |
+
"loss": 37.54802978515625,
|
| 143 |
+
"step": 950
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.08492478848419868,
|
| 147 |
+
"grad_norm": 184.31649780273438,
|
| 148 |
+
"learning_rate": 4.6548894263994475e-05,
|
| 149 |
+
"loss": 35.32251953125,
|
| 150 |
+
"step": 1000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.08917102790840861,
|
| 154 |
+
"grad_norm": 190.4438934326172,
|
| 155 |
+
"learning_rate": 4.633293020041466e-05,
|
| 156 |
+
"loss": 35.93584228515625,
|
| 157 |
+
"step": 1050
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.09341726733261856,
|
| 161 |
+
"grad_norm": 216.86279296875,
|
| 162 |
+
"learning_rate": 4.611696613683483e-05,
|
| 163 |
+
"loss": 34.81772216796875,
|
| 164 |
+
"step": 1100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.09766350675682849,
|
| 168 |
+
"grad_norm": 179.98419189453125,
|
| 169 |
+
"learning_rate": 4.590100207325501e-05,
|
| 170 |
+
"loss": 34.99229248046875,
|
| 171 |
+
"step": 1150
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.10190974618103842,
|
| 175 |
+
"grad_norm": 186.61732482910156,
|
| 176 |
+
"learning_rate": 4.568503800967519e-05,
|
| 177 |
+
"loss": 33.92795654296875,
|
| 178 |
+
"step": 1200
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.10615598560524835,
|
| 182 |
+
"grad_norm": 185.78001403808594,
|
| 183 |
+
"learning_rate": 4.546907394609537e-05,
|
| 184 |
+
"loss": 34.2297021484375,
|
| 185 |
+
"step": 1250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.11040222502945829,
|
| 189 |
+
"grad_norm": 236.7795867919922,
|
| 190 |
+
"learning_rate": 4.525310988251556e-05,
|
| 191 |
+
"loss": 33.76176513671875,
|
| 192 |
+
"step": 1300
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.11464846445366822,
|
| 196 |
+
"grad_norm": 210.53762817382812,
|
| 197 |
+
"learning_rate": 4.503714581893573e-05,
|
| 198 |
+
"loss": 32.79689453125,
|
| 199 |
+
"step": 1350
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.11889470387787815,
|
| 203 |
+
"grad_norm": 278.4184875488281,
|
| 204 |
+
"learning_rate": 4.4821181755355913e-05,
|
| 205 |
+
"loss": 31.72260009765625,
|
| 206 |
+
"step": 1400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.1231409433020881,
|
| 210 |
+
"grad_norm": 255.81129455566406,
|
| 211 |
+
"learning_rate": 4.460521769177609e-05,
|
| 212 |
+
"loss": 32.5959912109375,
|
| 213 |
+
"step": 1450
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.127387182726298,
|
| 217 |
+
"grad_norm": 188.4405975341797,
|
| 218 |
+
"learning_rate": 4.438925362819627e-05,
|
| 219 |
+
"loss": 31.72834716796875,
|
| 220 |
+
"step": 1500
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.13163342215050797,
|
| 224 |
+
"grad_norm": 196.35501098632812,
|
| 225 |
+
"learning_rate": 4.417328956461645e-05,
|
| 226 |
+
"loss": 32.69740478515625,
|
| 227 |
+
"step": 1550
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.1358796615747179,
|
| 231 |
+
"grad_norm": 186.20811462402344,
|
| 232 |
+
"learning_rate": 4.3957325501036626e-05,
|
| 233 |
+
"loss": 31.13796875,
|
| 234 |
+
"step": 1600
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.14012590099892783,
|
| 238 |
+
"grad_norm": 158.75543212890625,
|
| 239 |
+
"learning_rate": 4.374136143745681e-05,
|
| 240 |
+
"loss": 32.08728271484375,
|
| 241 |
+
"step": 1650
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.14437214042313776,
|
| 245 |
+
"grad_norm": 207.9669952392578,
|
| 246 |
+
"learning_rate": 4.352539737387699e-05,
|
| 247 |
+
"loss": 31.42122802734375,
|
| 248 |
+
"step": 1700
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.1486183798473477,
|
| 252 |
+
"grad_norm": 184.7431640625,
|
| 253 |
+
"learning_rate": 4.330943331029717e-05,
|
| 254 |
+
"loss": 31.1582763671875,
|
| 255 |
+
"step": 1750
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.15286461927155762,
|
| 259 |
+
"grad_norm": 155.47659301757812,
|
| 260 |
+
"learning_rate": 4.309346924671735e-05,
|
| 261 |
+
"loss": 30.77266845703125,
|
| 262 |
+
"step": 1800
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.15711085869576757,
|
| 266 |
+
"grad_norm": 163.03013610839844,
|
| 267 |
+
"learning_rate": 4.2877505183137527e-05,
|
| 268 |
+
"loss": 30.72901123046875,
|
| 269 |
+
"step": 1850
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.1613570981199775,
|
| 273 |
+
"grad_norm": 240.4451904296875,
|
| 274 |
+
"learning_rate": 4.266154111955771e-05,
|
| 275 |
+
"loss": 30.06856689453125,
|
| 276 |
+
"step": 1900
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.16560333754418743,
|
| 280 |
+
"grad_norm": 193.2200164794922,
|
| 281 |
+
"learning_rate": 4.244557705597789e-05,
|
| 282 |
+
"loss": 29.64366455078125,
|
| 283 |
+
"step": 1950
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.16984957696839736,
|
| 287 |
+
"grad_norm": 197.7560577392578,
|
| 288 |
+
"learning_rate": 4.2229612992398064e-05,
|
| 289 |
+
"loss": 29.403779296875,
|
| 290 |
+
"step": 2000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.1740958163926073,
|
| 294 |
+
"grad_norm": 172.2969512939453,
|
| 295 |
+
"learning_rate": 4.2013648928818246e-05,
|
| 296 |
+
"loss": 29.76959716796875,
|
| 297 |
+
"step": 2050
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.17834205581681722,
|
| 301 |
+
"grad_norm": 171.65423583984375,
|
| 302 |
+
"learning_rate": 4.179768486523843e-05,
|
| 303 |
+
"loss": 28.84666259765625,
|
| 304 |
+
"step": 2100
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.18258829524102715,
|
| 308 |
+
"grad_norm": 253.046875,
|
| 309 |
+
"learning_rate": 4.158172080165861e-05,
|
| 310 |
+
"loss": 28.4874462890625,
|
| 311 |
+
"step": 2150
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.1868345346652371,
|
| 315 |
+
"grad_norm": 174.5941162109375,
|
| 316 |
+
"learning_rate": 4.136575673807879e-05,
|
| 317 |
+
"loss": 28.8812744140625,
|
| 318 |
+
"step": 2200
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.19108077408944704,
|
| 322 |
+
"grad_norm": 187.45791625976562,
|
| 323 |
+
"learning_rate": 4.1149792674498965e-05,
|
| 324 |
+
"loss": 29.3225390625,
|
| 325 |
+
"step": 2250
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.19532701351365697,
|
| 329 |
+
"grad_norm": 158.6984405517578,
|
| 330 |
+
"learning_rate": 4.0933828610919146e-05,
|
| 331 |
+
"loss": 29.0456494140625,
|
| 332 |
+
"step": 2300
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.1995732529378669,
|
| 336 |
+
"grad_norm": 162.81512451171875,
|
| 337 |
+
"learning_rate": 4.071786454733932e-05,
|
| 338 |
+
"loss": 29.16901123046875,
|
| 339 |
+
"step": 2350
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.20381949236207683,
|
| 343 |
+
"grad_norm": 201.16639709472656,
|
| 344 |
+
"learning_rate": 4.05019004837595e-05,
|
| 345 |
+
"loss": 29.002841796875,
|
| 346 |
+
"step": 2400
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.20806573178628676,
|
| 350 |
+
"grad_norm": 185.6593475341797,
|
| 351 |
+
"learning_rate": 4.0285936420179684e-05,
|
| 352 |
+
"loss": 28.537255859375,
|
| 353 |
+
"step": 2450
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.2123119712104967,
|
| 357 |
+
"grad_norm": 166.38766479492188,
|
| 358 |
+
"learning_rate": 4.0069972356599866e-05,
|
| 359 |
+
"loss": 27.69727294921875,
|
| 360 |
+
"step": 2500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.21655821063470665,
|
| 364 |
+
"grad_norm": 174.38104248046875,
|
| 365 |
+
"learning_rate": 3.985400829302005e-05,
|
| 366 |
+
"loss": 28.5172802734375,
|
| 367 |
+
"step": 2550
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.22080445005891658,
|
| 371 |
+
"grad_norm": 149.02406311035156,
|
| 372 |
+
"learning_rate": 3.963804422944022e-05,
|
| 373 |
+
"loss": 28.4050244140625,
|
| 374 |
+
"step": 2600
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.2250506894831265,
|
| 378 |
+
"grad_norm": 192.12002563476562,
|
| 379 |
+
"learning_rate": 3.94220801658604e-05,
|
| 380 |
+
"loss": 26.98137451171875,
|
| 381 |
+
"step": 2650
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.22929692890733644,
|
| 385 |
+
"grad_norm": 170.00814819335938,
|
| 386 |
+
"learning_rate": 3.9206116102280585e-05,
|
| 387 |
+
"loss": 26.99474853515625,
|
| 388 |
+
"step": 2700
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.23354316833154637,
|
| 392 |
+
"grad_norm": 145.735595703125,
|
| 393 |
+
"learning_rate": 3.899015203870076e-05,
|
| 394 |
+
"loss": 27.34177001953125,
|
| 395 |
+
"step": 2750
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.2377894077557563,
|
| 399 |
+
"grad_norm": 169.86480712890625,
|
| 400 |
+
"learning_rate": 3.877418797512094e-05,
|
| 401 |
+
"loss": 27.52475830078125,
|
| 402 |
+
"step": 2800
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.24203564717996623,
|
| 406 |
+
"grad_norm": 135.18838500976562,
|
| 407 |
+
"learning_rate": 3.855822391154112e-05,
|
| 408 |
+
"loss": 27.2542626953125,
|
| 409 |
+
"step": 2850
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.2462818866041762,
|
| 413 |
+
"grad_norm": 149.96510314941406,
|
| 414 |
+
"learning_rate": 3.8342259847961304e-05,
|
| 415 |
+
"loss": 27.07297119140625,
|
| 416 |
+
"step": 2900
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.2505281260283861,
|
| 420 |
+
"grad_norm": 159.20584106445312,
|
| 421 |
+
"learning_rate": 3.8126295784381485e-05,
|
| 422 |
+
"loss": 27.08953125,
|
| 423 |
+
"step": 2950
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.254774365452596,
|
| 427 |
+
"grad_norm": 136.2477569580078,
|
| 428 |
+
"learning_rate": 3.791033172080166e-05,
|
| 429 |
+
"loss": 26.437490234375,
|
| 430 |
+
"step": 3000
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.259020604876806,
|
| 434 |
+
"grad_norm": 164.90737915039062,
|
| 435 |
+
"learning_rate": 3.769436765722184e-05,
|
| 436 |
+
"loss": 26.1881103515625,
|
| 437 |
+
"step": 3050
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.26326684430101593,
|
| 441 |
+
"grad_norm": 154.27102661132812,
|
| 442 |
+
"learning_rate": 3.747840359364202e-05,
|
| 443 |
+
"loss": 26.475185546875,
|
| 444 |
+
"step": 3100
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.26751308372522586,
|
| 448 |
+
"grad_norm": 174.78500366210938,
|
| 449 |
+
"learning_rate": 3.72624395300622e-05,
|
| 450 |
+
"loss": 26.410517578125,
|
| 451 |
+
"step": 3150
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.2717593231494358,
|
| 455 |
+
"grad_norm": 167.15457153320312,
|
| 456 |
+
"learning_rate": 3.704647546648238e-05,
|
| 457 |
+
"loss": 25.28077880859375,
|
| 458 |
+
"step": 3200
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.2760055625736457,
|
| 462 |
+
"grad_norm": 182.3014373779297,
|
| 463 |
+
"learning_rate": 3.6830511402902554e-05,
|
| 464 |
+
"loss": 26.1413818359375,
|
| 465 |
+
"step": 3250
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.28025180199785565,
|
| 469 |
+
"grad_norm": 185.13711547851562,
|
| 470 |
+
"learning_rate": 3.661454733932274e-05,
|
| 471 |
+
"loss": 25.593623046875,
|
| 472 |
+
"step": 3300
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.2844980414220656,
|
| 476 |
+
"grad_norm": 136.2028045654297,
|
| 477 |
+
"learning_rate": 3.639858327574292e-05,
|
| 478 |
+
"loss": 26.10810791015625,
|
| 479 |
+
"step": 3350
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.2887442808462755,
|
| 483 |
+
"grad_norm": 160.10540771484375,
|
| 484 |
+
"learning_rate": 3.61826192121631e-05,
|
| 485 |
+
"loss": 26.09581298828125,
|
| 486 |
+
"step": 3400
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.29299052027048544,
|
| 490 |
+
"grad_norm": 170.12704467773438,
|
| 491 |
+
"learning_rate": 3.596665514858328e-05,
|
| 492 |
+
"loss": 26.3189306640625,
|
| 493 |
+
"step": 3450
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.2972367596946954,
|
| 497 |
+
"grad_norm": 154.6916961669922,
|
| 498 |
+
"learning_rate": 3.5750691085003455e-05,
|
| 499 |
+
"loss": 25.707509765625,
|
| 500 |
+
"step": 3500
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.3014829991189053,
|
| 504 |
+
"grad_norm": 175.77297973632812,
|
| 505 |
+
"learning_rate": 3.5534727021423636e-05,
|
| 506 |
+
"loss": 26.1056396484375,
|
| 507 |
+
"step": 3550
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.30572923854311523,
|
| 511 |
+
"grad_norm": 143.42288208007812,
|
| 512 |
+
"learning_rate": 3.531876295784382e-05,
|
| 513 |
+
"loss": 25.2949560546875,
|
| 514 |
+
"step": 3600
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.30997547796732516,
|
| 518 |
+
"grad_norm": 171.7968292236328,
|
| 519 |
+
"learning_rate": 3.510279889426399e-05,
|
| 520 |
+
"loss": 25.76105712890625,
|
| 521 |
+
"step": 3650
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.31422171739153515,
|
| 525 |
+
"grad_norm": 165.98411560058594,
|
| 526 |
+
"learning_rate": 3.488683483068418e-05,
|
| 527 |
+
"loss": 25.60622802734375,
|
| 528 |
+
"step": 3700
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.3184679568157451,
|
| 532 |
+
"grad_norm": 157.76429748535156,
|
| 533 |
+
"learning_rate": 3.4670870767104355e-05,
|
| 534 |
+
"loss": 24.7918017578125,
|
| 535 |
+
"step": 3750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.322714196239955,
|
| 539 |
+
"grad_norm": 168.3741455078125,
|
| 540 |
+
"learning_rate": 3.445490670352454e-05,
|
| 541 |
+
"loss": 24.83272705078125,
|
| 542 |
+
"step": 3800
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.32696043566416494,
|
| 546 |
+
"grad_norm": 165.90496826171875,
|
| 547 |
+
"learning_rate": 3.423894263994472e-05,
|
| 548 |
+
"loss": 26.2589599609375,
|
| 549 |
+
"step": 3850
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.33120667508837487,
|
| 553 |
+
"grad_norm": 142.2030792236328,
|
| 554 |
+
"learning_rate": 3.402297857636489e-05,
|
| 555 |
+
"loss": 24.28029541015625,
|
| 556 |
+
"step": 3900
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.3354529145125848,
|
| 560 |
+
"grad_norm": 159.05149841308594,
|
| 561 |
+
"learning_rate": 3.3807014512785074e-05,
|
| 562 |
+
"loss": 24.80717041015625,
|
| 563 |
+
"step": 3950
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.33969915393679473,
|
| 567 |
+
"grad_norm": 138.15611267089844,
|
| 568 |
+
"learning_rate": 3.359105044920525e-05,
|
| 569 |
+
"loss": 24.88008056640625,
|
| 570 |
+
"step": 4000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.34394539336100466,
|
| 574 |
+
"grad_norm": 196.62498474121094,
|
| 575 |
+
"learning_rate": 3.337508638562543e-05,
|
| 576 |
+
"loss": 24.37576171875,
|
| 577 |
+
"step": 4050
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.3481916327852146,
|
| 581 |
+
"grad_norm": 155.42593383789062,
|
| 582 |
+
"learning_rate": 3.315912232204562e-05,
|
| 583 |
+
"loss": 25.1333447265625,
|
| 584 |
+
"step": 4100
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.3524378722094245,
|
| 588 |
+
"grad_norm": 146.24502563476562,
|
| 589 |
+
"learning_rate": 3.2943158258465794e-05,
|
| 590 |
+
"loss": 23.95904296875,
|
| 591 |
+
"step": 4150
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.35668411163363445,
|
| 595 |
+
"grad_norm": 140.36927795410156,
|
| 596 |
+
"learning_rate": 3.2727194194885975e-05,
|
| 597 |
+
"loss": 24.21125,
|
| 598 |
+
"step": 4200
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.3609303510578444,
|
| 602 |
+
"grad_norm": 151.30316162109375,
|
| 603 |
+
"learning_rate": 3.251123013130615e-05,
|
| 604 |
+
"loss": 24.09343505859375,
|
| 605 |
+
"step": 4250
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.3651765904820543,
|
| 609 |
+
"grad_norm": 209.92578125,
|
| 610 |
+
"learning_rate": 3.229526606772633e-05,
|
| 611 |
+
"loss": 23.6073779296875,
|
| 612 |
+
"step": 4300
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.36942282990626424,
|
| 616 |
+
"grad_norm": 162.1195526123047,
|
| 617 |
+
"learning_rate": 3.207930200414651e-05,
|
| 618 |
+
"loss": 24.49541748046875,
|
| 619 |
+
"step": 4350
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.3736690693304742,
|
| 623 |
+
"grad_norm": 170.38694763183594,
|
| 624 |
+
"learning_rate": 3.186333794056669e-05,
|
| 625 |
+
"loss": 24.1321630859375,
|
| 626 |
+
"step": 4400
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.37791530875468415,
|
| 630 |
+
"grad_norm": 161.4326934814453,
|
| 631 |
+
"learning_rate": 3.164737387698687e-05,
|
| 632 |
+
"loss": 23.7575341796875,
|
| 633 |
+
"step": 4450
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.3821615481788941,
|
| 637 |
+
"grad_norm": 171.4528045654297,
|
| 638 |
+
"learning_rate": 3.143140981340705e-05,
|
| 639 |
+
"loss": 24.50120849609375,
|
| 640 |
+
"step": 4500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.386407787603104,
|
| 644 |
+
"grad_norm": 160.71112060546875,
|
| 645 |
+
"learning_rate": 3.121544574982723e-05,
|
| 646 |
+
"loss": 23.270244140625,
|
| 647 |
+
"step": 4550
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.39065402702731394,
|
| 651 |
+
"grad_norm": 139.88951110839844,
|
| 652 |
+
"learning_rate": 3.099948168624741e-05,
|
| 653 |
+
"loss": 24.25181640625,
|
| 654 |
+
"step": 4600
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.3949002664515239,
|
| 658 |
+
"grad_norm": 152.9877166748047,
|
| 659 |
+
"learning_rate": 3.078351762266759e-05,
|
| 660 |
+
"loss": 23.2979345703125,
|
| 661 |
+
"step": 4650
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.3991465058757338,
|
| 665 |
+
"grad_norm": 179.30581665039062,
|
| 666 |
+
"learning_rate": 3.056755355908777e-05,
|
| 667 |
+
"loss": 24.148583984375,
|
| 668 |
+
"step": 4700
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.40339274529994373,
|
| 672 |
+
"grad_norm": 143.60121154785156,
|
| 673 |
+
"learning_rate": 3.0351589495507948e-05,
|
| 674 |
+
"loss": 24.34101318359375,
|
| 675 |
+
"step": 4750
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.40763898472415366,
|
| 679 |
+
"grad_norm": 133.65771484375,
|
| 680 |
+
"learning_rate": 3.0135625431928126e-05,
|
| 681 |
+
"loss": 23.299140625,
|
| 682 |
+
"step": 4800
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.4118852241483636,
|
| 686 |
+
"grad_norm": 162.5352783203125,
|
| 687 |
+
"learning_rate": 2.9919661368348307e-05,
|
| 688 |
+
"loss": 23.08830322265625,
|
| 689 |
+
"step": 4850
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.4161314635725735,
|
| 693 |
+
"grad_norm": 158.4131317138672,
|
| 694 |
+
"learning_rate": 2.9703697304768485e-05,
|
| 695 |
+
"loss": 22.8610205078125,
|
| 696 |
+
"step": 4900
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.42037770299678345,
|
| 700 |
+
"grad_norm": 124.68849182128906,
|
| 701 |
+
"learning_rate": 2.948773324118867e-05,
|
| 702 |
+
"loss": 23.79708740234375,
|
| 703 |
+
"step": 4950
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.4246239424209934,
|
| 707 |
+
"grad_norm": 163.76657104492188,
|
| 708 |
+
"learning_rate": 2.9271769177608848e-05,
|
| 709 |
+
"loss": 23.52044189453125,
|
| 710 |
+
"step": 5000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.42887018184520337,
|
| 714 |
+
"grad_norm": 187.97413635253906,
|
| 715 |
+
"learning_rate": 2.9055805114029026e-05,
|
| 716 |
+
"loss": 24.02717529296875,
|
| 717 |
+
"step": 5050
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.4331164212694133,
|
| 721 |
+
"grad_norm": 154.0001220703125,
|
| 722 |
+
"learning_rate": 2.8839841050449208e-05,
|
| 723 |
+
"loss": 23.41170166015625,
|
| 724 |
+
"step": 5100
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.43736266069362323,
|
| 728 |
+
"grad_norm": 141.18080139160156,
|
| 729 |
+
"learning_rate": 2.8623876986869386e-05,
|
| 730 |
+
"loss": 23.554814453125,
|
| 731 |
+
"step": 5150
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.44160890011783316,
|
| 735 |
+
"grad_norm": 158.98216247558594,
|
| 736 |
+
"learning_rate": 2.8407912923289564e-05,
|
| 737 |
+
"loss": 23.31008544921875,
|
| 738 |
+
"step": 5200
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.4458551395420431,
|
| 742 |
+
"grad_norm": 146.218994140625,
|
| 743 |
+
"learning_rate": 2.8191948859709742e-05,
|
| 744 |
+
"loss": 23.71553955078125,
|
| 745 |
+
"step": 5250
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.450101378966253,
|
| 749 |
+
"grad_norm": 156.51646423339844,
|
| 750 |
+
"learning_rate": 2.7975984796129924e-05,
|
| 751 |
+
"loss": 22.9737060546875,
|
| 752 |
+
"step": 5300
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.45434761839046295,
|
| 756 |
+
"grad_norm": 144.48443603515625,
|
| 757 |
+
"learning_rate": 2.776002073255011e-05,
|
| 758 |
+
"loss": 24.7643798828125,
|
| 759 |
+
"step": 5350
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.4585938578146729,
|
| 763 |
+
"grad_norm": 160.95411682128906,
|
| 764 |
+
"learning_rate": 2.7544056668970287e-05,
|
| 765 |
+
"loss": 23.05649658203125,
|
| 766 |
+
"step": 5400
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.4628400972388828,
|
| 770 |
+
"grad_norm": 152.6392059326172,
|
| 771 |
+
"learning_rate": 2.7328092605390465e-05,
|
| 772 |
+
"loss": 23.15297119140625,
|
| 773 |
+
"step": 5450
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.46708633666309274,
|
| 777 |
+
"grad_norm": 190.5279083251953,
|
| 778 |
+
"learning_rate": 2.7112128541810643e-05,
|
| 779 |
+
"loss": 21.804638671875,
|
| 780 |
+
"step": 5500
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.47133257608730267,
|
| 784 |
+
"grad_norm": 170.51712036132812,
|
| 785 |
+
"learning_rate": 2.6896164478230824e-05,
|
| 786 |
+
"loss": 22.71973388671875,
|
| 787 |
+
"step": 5550
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.4755788155115126,
|
| 791 |
+
"grad_norm": 151.53326416015625,
|
| 792 |
+
"learning_rate": 2.6680200414651002e-05,
|
| 793 |
+
"loss": 22.90906494140625,
|
| 794 |
+
"step": 5600
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.47982505493572253,
|
| 798 |
+
"grad_norm": 147.06654357910156,
|
| 799 |
+
"learning_rate": 2.646423635107118e-05,
|
| 800 |
+
"loss": 24.11918701171875,
|
| 801 |
+
"step": 5650
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.48407129435993246,
|
| 805 |
+
"grad_norm": 167.34326171875,
|
| 806 |
+
"learning_rate": 2.624827228749136e-05,
|
| 807 |
+
"loss": 22.7313134765625,
|
| 808 |
+
"step": 5700
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.48831753378414244,
|
| 812 |
+
"grad_norm": 136.9669647216797,
|
| 813 |
+
"learning_rate": 2.6032308223911543e-05,
|
| 814 |
+
"loss": 21.70623046875,
|
| 815 |
+
"step": 5750
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.4925637732083524,
|
| 819 |
+
"grad_norm": 159.77626037597656,
|
| 820 |
+
"learning_rate": 2.5816344160331725e-05,
|
| 821 |
+
"loss": 22.6339453125,
|
| 822 |
+
"step": 5800
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.4968100126325623,
|
| 826 |
+
"grad_norm": 198.8340301513672,
|
| 827 |
+
"learning_rate": 2.5600380096751903e-05,
|
| 828 |
+
"loss": 22.50951171875,
|
| 829 |
+
"step": 5850
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.5010562520567722,
|
| 833 |
+
"grad_norm": 139.2235870361328,
|
| 834 |
+
"learning_rate": 2.538441603317208e-05,
|
| 835 |
+
"loss": 22.05862548828125,
|
| 836 |
+
"step": 5900
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.5053024914809822,
|
| 840 |
+
"grad_norm": 145.6766815185547,
|
| 841 |
+
"learning_rate": 2.516845196959226e-05,
|
| 842 |
+
"loss": 23.03335205078125,
|
| 843 |
+
"step": 5950
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.509548730905192,
|
| 847 |
+
"grad_norm": 159.59835815429688,
|
| 848 |
+
"learning_rate": 2.495248790601244e-05,
|
| 849 |
+
"loss": 23.1497705078125,
|
| 850 |
+
"step": 6000
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.513794970329402,
|
| 854 |
+
"grad_norm": 200.4053497314453,
|
| 855 |
+
"learning_rate": 2.4736523842432622e-05,
|
| 856 |
+
"loss": 22.751318359375,
|
| 857 |
+
"step": 6050
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.518041209753612,
|
| 861 |
+
"grad_norm": 146.4293670654297,
|
| 862 |
+
"learning_rate": 2.45205597788528e-05,
|
| 863 |
+
"loss": 22.441064453125,
|
| 864 |
+
"step": 6100
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.5222874491778219,
|
| 868 |
+
"grad_norm": 140.7433319091797,
|
| 869 |
+
"learning_rate": 2.430459571527298e-05,
|
| 870 |
+
"loss": 21.8677880859375,
|
| 871 |
+
"step": 6150
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.5265336886020319,
|
| 875 |
+
"grad_norm": 145.3983154296875,
|
| 876 |
+
"learning_rate": 2.4088631651693156e-05,
|
| 877 |
+
"loss": 21.99537109375,
|
| 878 |
+
"step": 6200
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.5307799280262417,
|
| 882 |
+
"grad_norm": 171.57505798339844,
|
| 883 |
+
"learning_rate": 2.387266758811334e-05,
|
| 884 |
+
"loss": 22.0473974609375,
|
| 885 |
+
"step": 6250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.5350261674504517,
|
| 889 |
+
"grad_norm": 190.87144470214844,
|
| 890 |
+
"learning_rate": 2.365670352453352e-05,
|
| 891 |
+
"loss": 22.20819580078125,
|
| 892 |
+
"step": 6300
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.5392724068746616,
|
| 896 |
+
"grad_norm": 131.15647888183594,
|
| 897 |
+
"learning_rate": 2.3440739460953698e-05,
|
| 898 |
+
"loss": 22.3381689453125,
|
| 899 |
+
"step": 6350
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.5435186462988716,
|
| 903 |
+
"grad_norm": 130.159423828125,
|
| 904 |
+
"learning_rate": 2.3224775397373876e-05,
|
| 905 |
+
"loss": 21.13576171875,
|
| 906 |
+
"step": 6400
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.5477648857230815,
|
| 910 |
+
"grad_norm": 153.32504272460938,
|
| 911 |
+
"learning_rate": 2.3008811333794057e-05,
|
| 912 |
+
"loss": 21.22865478515625,
|
| 913 |
+
"step": 6450
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.5520111251472914,
|
| 917 |
+
"grad_norm": 137.8641357421875,
|
| 918 |
+
"learning_rate": 2.279284727021424e-05,
|
| 919 |
+
"loss": 21.42198974609375,
|
| 920 |
+
"step": 6500
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.5562573645715013,
|
| 924 |
+
"grad_norm": 152.36447143554688,
|
| 925 |
+
"learning_rate": 2.2576883206634417e-05,
|
| 926 |
+
"loss": 22.49668212890625,
|
| 927 |
+
"step": 6550
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.5605036039957113,
|
| 931 |
+
"grad_norm": 170.3688507080078,
|
| 932 |
+
"learning_rate": 2.2360919143054595e-05,
|
| 933 |
+
"loss": 21.45054443359375,
|
| 934 |
+
"step": 6600
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.5647498434199212,
|
| 938 |
+
"grad_norm": 119.01740264892578,
|
| 939 |
+
"learning_rate": 2.2144955079474776e-05,
|
| 940 |
+
"loss": 22.2169287109375,
|
| 941 |
+
"step": 6650
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.5689960828441312,
|
| 945 |
+
"grad_norm": 147.85865783691406,
|
| 946 |
+
"learning_rate": 2.1928991015894958e-05,
|
| 947 |
+
"loss": 22.83664794921875,
|
| 948 |
+
"step": 6700
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.5732423222683412,
|
| 952 |
+
"grad_norm": 153.3690948486328,
|
| 953 |
+
"learning_rate": 2.1713026952315136e-05,
|
| 954 |
+
"loss": 21.9330712890625,
|
| 955 |
+
"step": 6750
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.577488561692551,
|
| 959 |
+
"grad_norm": 137.37164306640625,
|
| 960 |
+
"learning_rate": 2.1497062888735314e-05,
|
| 961 |
+
"loss": 20.79832763671875,
|
| 962 |
+
"step": 6800
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.581734801116761,
|
| 966 |
+
"grad_norm": 130.9550018310547,
|
| 967 |
+
"learning_rate": 2.1281098825155495e-05,
|
| 968 |
+
"loss": 21.09296630859375,
|
| 969 |
+
"step": 6850
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.5859810405409709,
|
| 973 |
+
"grad_norm": 121.02996063232422,
|
| 974 |
+
"learning_rate": 2.1065134761575674e-05,
|
| 975 |
+
"loss": 21.2236083984375,
|
| 976 |
+
"step": 6900
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.5902272799651809,
|
| 980 |
+
"grad_norm": 169.94479370117188,
|
| 981 |
+
"learning_rate": 2.0849170697995855e-05,
|
| 982 |
+
"loss": 21.0869140625,
|
| 983 |
+
"step": 6950
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.5944735193893907,
|
| 987 |
+
"grad_norm": 139.7215576171875,
|
| 988 |
+
"learning_rate": 2.0633206634416033e-05,
|
| 989 |
+
"loss": 20.95746337890625,
|
| 990 |
+
"step": 7000
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.5987197588136007,
|
| 994 |
+
"grad_norm": 135.5031280517578,
|
| 995 |
+
"learning_rate": 2.0417242570836215e-05,
|
| 996 |
+
"loss": 21.83375,
|
| 997 |
+
"step": 7050
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.6029659982378106,
|
| 1001 |
+
"grad_norm": 143.67396545410156,
|
| 1002 |
+
"learning_rate": 2.0201278507256393e-05,
|
| 1003 |
+
"loss": 21.9516015625,
|
| 1004 |
+
"step": 7100
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.6072122376620206,
|
| 1008 |
+
"grad_norm": 242.10548400878906,
|
| 1009 |
+
"learning_rate": 1.9985314443676574e-05,
|
| 1010 |
+
"loss": 21.42175537109375,
|
| 1011 |
+
"step": 7150
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.6114584770862305,
|
| 1015 |
+
"grad_norm": 157.11415100097656,
|
| 1016 |
+
"learning_rate": 1.9769350380096752e-05,
|
| 1017 |
+
"loss": 21.08699951171875,
|
| 1018 |
+
"step": 7200
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.6157047165104405,
|
| 1022 |
+
"grad_norm": 124.51484680175781,
|
| 1023 |
+
"learning_rate": 1.9553386316516934e-05,
|
| 1024 |
+
"loss": 21.26003662109375,
|
| 1025 |
+
"step": 7250
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.6199509559346503,
|
| 1029 |
+
"grad_norm": 175.8770751953125,
|
| 1030 |
+
"learning_rate": 1.9337422252937112e-05,
|
| 1031 |
+
"loss": 21.37485107421875,
|
| 1032 |
+
"step": 7300
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.6241971953588603,
|
| 1036 |
+
"grad_norm": 139.2329559326172,
|
| 1037 |
+
"learning_rate": 1.912145818935729e-05,
|
| 1038 |
+
"loss": 21.006982421875,
|
| 1039 |
+
"step": 7350
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.6284434347830703,
|
| 1043 |
+
"grad_norm": 142.4539031982422,
|
| 1044 |
+
"learning_rate": 1.890549412577747e-05,
|
| 1045 |
+
"loss": 20.92181640625,
|
| 1046 |
+
"step": 7400
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.6326896742072802,
|
| 1050 |
+
"grad_norm": 143.5320281982422,
|
| 1051 |
+
"learning_rate": 1.8689530062197653e-05,
|
| 1052 |
+
"loss": 21.85162109375,
|
| 1053 |
+
"step": 7450
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.6369359136314902,
|
| 1057 |
+
"grad_norm": 152.91880798339844,
|
| 1058 |
+
"learning_rate": 1.847356599861783e-05,
|
| 1059 |
+
"loss": 20.92374755859375,
|
| 1060 |
+
"step": 7500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.6411821530557,
|
| 1064 |
+
"grad_norm": 162.99484252929688,
|
| 1065 |
+
"learning_rate": 1.825760193503801e-05,
|
| 1066 |
+
"loss": 21.4075537109375,
|
| 1067 |
+
"step": 7550
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.64542839247991,
|
| 1071 |
+
"grad_norm": 164.03475952148438,
|
| 1072 |
+
"learning_rate": 1.804163787145819e-05,
|
| 1073 |
+
"loss": 21.6792919921875,
|
| 1074 |
+
"step": 7600
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.6496746319041199,
|
| 1078 |
+
"grad_norm": 154.84764099121094,
|
| 1079 |
+
"learning_rate": 1.7825673807878372e-05,
|
| 1080 |
+
"loss": 20.593271484375,
|
| 1081 |
+
"step": 7650
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.6539208713283299,
|
| 1085 |
+
"grad_norm": 136.49598693847656,
|
| 1086 |
+
"learning_rate": 1.760970974429855e-05,
|
| 1087 |
+
"loss": 20.4891650390625,
|
| 1088 |
+
"step": 7700
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.6581671107525398,
|
| 1092 |
+
"grad_norm": 144.41871643066406,
|
| 1093 |
+
"learning_rate": 1.739374568071873e-05,
|
| 1094 |
+
"loss": 21.7032177734375,
|
| 1095 |
+
"step": 7750
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.6624133501767497,
|
| 1099 |
+
"grad_norm": 172.35324096679688,
|
| 1100 |
+
"learning_rate": 1.7177781617138906e-05,
|
| 1101 |
+
"loss": 21.29265380859375,
|
| 1102 |
+
"step": 7800
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.6666595896009596,
|
| 1106 |
+
"grad_norm": 154.5519256591797,
|
| 1107 |
+
"learning_rate": 1.6961817553559088e-05,
|
| 1108 |
+
"loss": 20.79191650390625,
|
| 1109 |
+
"step": 7850
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.6709058290251696,
|
| 1113 |
+
"grad_norm": 157.8460235595703,
|
| 1114 |
+
"learning_rate": 1.674585348997927e-05,
|
| 1115 |
+
"loss": 20.80486328125,
|
| 1116 |
+
"step": 7900
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.6751520684493795,
|
| 1120 |
+
"grad_norm": 160.06161499023438,
|
| 1121 |
+
"learning_rate": 1.6529889426399447e-05,
|
| 1122 |
+
"loss": 20.06711669921875,
|
| 1123 |
+
"step": 7950
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.6793983078735895,
|
| 1127 |
+
"grad_norm": 150.2727813720703,
|
| 1128 |
+
"learning_rate": 1.6313925362819626e-05,
|
| 1129 |
+
"loss": 20.79498046875,
|
| 1130 |
+
"step": 8000
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.6836445472977993,
|
| 1134 |
+
"grad_norm": 144.0926971435547,
|
| 1135 |
+
"learning_rate": 1.6097961299239807e-05,
|
| 1136 |
+
"loss": 21.199130859375,
|
| 1137 |
+
"step": 8050
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.6878907867220093,
|
| 1141 |
+
"grad_norm": 149.16781616210938,
|
| 1142 |
+
"learning_rate": 1.588199723565999e-05,
|
| 1143 |
+
"loss": 20.359921875,
|
| 1144 |
+
"step": 8100
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.6921370261462193,
|
| 1148 |
+
"grad_norm": 167.05862426757812,
|
| 1149 |
+
"learning_rate": 1.5666033172080167e-05,
|
| 1150 |
+
"loss": 20.74350341796875,
|
| 1151 |
+
"step": 8150
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.6963832655704292,
|
| 1155 |
+
"grad_norm": 164.15008544921875,
|
| 1156 |
+
"learning_rate": 1.5450069108500345e-05,
|
| 1157 |
+
"loss": 19.6603076171875,
|
| 1158 |
+
"step": 8200
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.7006295049946392,
|
| 1162 |
+
"grad_norm": 175.12493896484375,
|
| 1163 |
+
"learning_rate": 1.5234105044920526e-05,
|
| 1164 |
+
"loss": 20.55838623046875,
|
| 1165 |
+
"step": 8250
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.704875744418849,
|
| 1169 |
+
"grad_norm": 146.61241149902344,
|
| 1170 |
+
"learning_rate": 1.5018140981340706e-05,
|
| 1171 |
+
"loss": 20.462938232421877,
|
| 1172 |
+
"step": 8300
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.709121983843059,
|
| 1176 |
+
"grad_norm": 132.65420532226562,
|
| 1177 |
+
"learning_rate": 1.4802176917760884e-05,
|
| 1178 |
+
"loss": 20.41062255859375,
|
| 1179 |
+
"step": 8350
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.7133682232672689,
|
| 1183 |
+
"grad_norm": 145.90028381347656,
|
| 1184 |
+
"learning_rate": 1.4586212854181064e-05,
|
| 1185 |
+
"loss": 20.371319580078126,
|
| 1186 |
+
"step": 8400
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.7176144626914789,
|
| 1190 |
+
"grad_norm": 147.31033325195312,
|
| 1191 |
+
"learning_rate": 1.4370248790601245e-05,
|
| 1192 |
+
"loss": 20.66120361328125,
|
| 1193 |
+
"step": 8450
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.7218607021156888,
|
| 1197 |
+
"grad_norm": 136.2473602294922,
|
| 1198 |
+
"learning_rate": 1.4154284727021425e-05,
|
| 1199 |
+
"loss": 20.56037109375,
|
| 1200 |
+
"step": 8500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.7261069415398987,
|
| 1204 |
+
"grad_norm": 152.7163848876953,
|
| 1205 |
+
"learning_rate": 1.3938320663441603e-05,
|
| 1206 |
+
"loss": 20.41483154296875,
|
| 1207 |
+
"step": 8550
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.7303531809641086,
|
| 1211 |
+
"grad_norm": 144.92152404785156,
|
| 1212 |
+
"learning_rate": 1.3722356599861783e-05,
|
| 1213 |
+
"loss": 21.3026416015625,
|
| 1214 |
+
"step": 8600
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.7345994203883186,
|
| 1218 |
+
"grad_norm": 146.7015380859375,
|
| 1219 |
+
"learning_rate": 1.3506392536281965e-05,
|
| 1220 |
+
"loss": 20.11982177734375,
|
| 1221 |
+
"step": 8650
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.7388456598125285,
|
| 1225 |
+
"grad_norm": 122.79088592529297,
|
| 1226 |
+
"learning_rate": 1.3290428472702143e-05,
|
| 1227 |
+
"loss": 20.265406494140624,
|
| 1228 |
+
"step": 8700
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.7430918992367385,
|
| 1232 |
+
"grad_norm": 124.77399444580078,
|
| 1233 |
+
"learning_rate": 1.3074464409122322e-05,
|
| 1234 |
+
"loss": 20.56513427734375,
|
| 1235 |
+
"step": 8750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.7473381386609484,
|
| 1239 |
+
"grad_norm": 131.1453094482422,
|
| 1240 |
+
"learning_rate": 1.28585003455425e-05,
|
| 1241 |
+
"loss": 19.999232177734374,
|
| 1242 |
+
"step": 8800
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.7515843780851583,
|
| 1246 |
+
"grad_norm": 143.78489685058594,
|
| 1247 |
+
"learning_rate": 1.2642536281962684e-05,
|
| 1248 |
+
"loss": 20.73391845703125,
|
| 1249 |
+
"step": 8850
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.7558306175093683,
|
| 1253 |
+
"grad_norm": 144.25245666503906,
|
| 1254 |
+
"learning_rate": 1.2426572218382862e-05,
|
| 1255 |
+
"loss": 20.422523193359375,
|
| 1256 |
+
"step": 8900
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.7600768569335782,
|
| 1260 |
+
"grad_norm": 173.52023315429688,
|
| 1261 |
+
"learning_rate": 1.2210608154803042e-05,
|
| 1262 |
+
"loss": 19.5886962890625,
|
| 1263 |
+
"step": 8950
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.7643230963577882,
|
| 1267 |
+
"grad_norm": 138.1155242919922,
|
| 1268 |
+
"learning_rate": 1.1994644091223221e-05,
|
| 1269 |
+
"loss": 19.826878662109376,
|
| 1270 |
+
"step": 9000
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.768569335781998,
|
| 1274 |
+
"grad_norm": 141.95608520507812,
|
| 1275 |
+
"learning_rate": 1.17786800276434e-05,
|
| 1276 |
+
"loss": 20.6840185546875,
|
| 1277 |
+
"step": 9050
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.772815575206208,
|
| 1281 |
+
"grad_norm": 130.37266540527344,
|
| 1282 |
+
"learning_rate": 1.1562715964063581e-05,
|
| 1283 |
+
"loss": 20.7981884765625,
|
| 1284 |
+
"step": 9100
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.7770618146304179,
|
| 1288 |
+
"grad_norm": 129.22499084472656,
|
| 1289 |
+
"learning_rate": 1.1346751900483759e-05,
|
| 1290 |
+
"loss": 20.298045654296875,
|
| 1291 |
+
"step": 9150
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.7813080540546279,
|
| 1295 |
+
"grad_norm": 196.2136993408203,
|
| 1296 |
+
"learning_rate": 1.113078783690394e-05,
|
| 1297 |
+
"loss": 20.010155029296875,
|
| 1298 |
+
"step": 9200
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.7855542934788378,
|
| 1302 |
+
"grad_norm": 143.95925903320312,
|
| 1303 |
+
"learning_rate": 1.0914823773324119e-05,
|
| 1304 |
+
"loss": 20.69479248046875,
|
| 1305 |
+
"step": 9250
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.7898005329030477,
|
| 1309 |
+
"grad_norm": 140.12205505371094,
|
| 1310 |
+
"learning_rate": 1.06988597097443e-05,
|
| 1311 |
+
"loss": 20.101761474609376,
|
| 1312 |
+
"step": 9300
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.7940467723272576,
|
| 1316 |
+
"grad_norm": 148.62911987304688,
|
| 1317 |
+
"learning_rate": 1.0482895646164478e-05,
|
| 1318 |
+
"loss": 20.48222900390625,
|
| 1319 |
+
"step": 9350
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.7982930117514676,
|
| 1323 |
+
"grad_norm": 134.18821716308594,
|
| 1324 |
+
"learning_rate": 1.0266931582584658e-05,
|
| 1325 |
+
"loss": 21.5114208984375,
|
| 1326 |
+
"step": 9400
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.8025392511756776,
|
| 1330 |
+
"grad_norm": 112.08241271972656,
|
| 1331 |
+
"learning_rate": 1.0050967519004838e-05,
|
| 1332 |
+
"loss": 19.884307861328125,
|
| 1333 |
+
"step": 9450
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.8067854905998875,
|
| 1337 |
+
"grad_norm": 130.55465698242188,
|
| 1338 |
+
"learning_rate": 9.835003455425018e-06,
|
| 1339 |
+
"loss": 19.306004638671876,
|
| 1340 |
+
"step": 9500
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.8110317300240975,
|
| 1344 |
+
"grad_norm": 175.3483428955078,
|
| 1345 |
+
"learning_rate": 9.619039391845197e-06,
|
| 1346 |
+
"loss": 20.79504638671875,
|
| 1347 |
+
"step": 9550
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.8152779694483073,
|
| 1351 |
+
"grad_norm": 123.85011291503906,
|
| 1352 |
+
"learning_rate": 9.403075328265377e-06,
|
| 1353 |
+
"loss": 19.556348876953123,
|
| 1354 |
+
"step": 9600
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.8195242088725173,
|
| 1358 |
+
"grad_norm": 140.8203125,
|
| 1359 |
+
"learning_rate": 9.187111264685557e-06,
|
| 1360 |
+
"loss": 20.43146484375,
|
| 1361 |
+
"step": 9650
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.8237704482967272,
|
| 1365 |
+
"grad_norm": 136.66036987304688,
|
| 1366 |
+
"learning_rate": 8.971147201105737e-06,
|
| 1367 |
+
"loss": 20.33073486328125,
|
| 1368 |
+
"step": 9700
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.8280166877209372,
|
| 1372 |
+
"grad_norm": 146.3016815185547,
|
| 1373 |
+
"learning_rate": 8.755183137525917e-06,
|
| 1374 |
+
"loss": 19.646549072265625,
|
| 1375 |
+
"step": 9750
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.832262927145147,
|
| 1379 |
+
"grad_norm": 148.30975341796875,
|
| 1380 |
+
"learning_rate": 8.539219073946096e-06,
|
| 1381 |
+
"loss": 19.05823486328125,
|
| 1382 |
+
"step": 9800
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.836509166569357,
|
| 1386 |
+
"grad_norm": 140.53770446777344,
|
| 1387 |
+
"learning_rate": 8.323255010366274e-06,
|
| 1388 |
+
"loss": 19.65471923828125,
|
| 1389 |
+
"step": 9850
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.8407554059935669,
|
| 1393 |
+
"grad_norm": 137.27951049804688,
|
| 1394 |
+
"learning_rate": 8.107290946786456e-06,
|
| 1395 |
+
"loss": 20.391009521484374,
|
| 1396 |
+
"step": 9900
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.8450016454177769,
|
| 1400 |
+
"grad_norm": 151.11325073242188,
|
| 1401 |
+
"learning_rate": 7.891326883206634e-06,
|
| 1402 |
+
"loss": 19.69550048828125,
|
| 1403 |
+
"step": 9950
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.8492478848419868,
|
| 1407 |
+
"grad_norm": 137.7413787841797,
|
| 1408 |
+
"learning_rate": 7.675362819626816e-06,
|
| 1409 |
+
"loss": 19.3516943359375,
|
| 1410 |
+
"step": 10000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.8534941242661968,
|
| 1414 |
+
"grad_norm": 112.80725860595703,
|
| 1415 |
+
"learning_rate": 7.459398756046994e-06,
|
| 1416 |
+
"loss": 19.201815185546874,
|
| 1417 |
+
"step": 10050
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.8577403636904067,
|
| 1421 |
+
"grad_norm": 150.98226928710938,
|
| 1422 |
+
"learning_rate": 7.243434692467174e-06,
|
| 1423 |
+
"loss": 19.6190673828125,
|
| 1424 |
+
"step": 10100
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.8619866031146166,
|
| 1428 |
+
"grad_norm": 151.21493530273438,
|
| 1429 |
+
"learning_rate": 7.027470628887353e-06,
|
| 1430 |
+
"loss": 20.10290771484375,
|
| 1431 |
+
"step": 10150
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.8662328425388266,
|
| 1435 |
+
"grad_norm": 166.5272674560547,
|
| 1436 |
+
"learning_rate": 6.811506565307534e-06,
|
| 1437 |
+
"loss": 20.10521728515625,
|
| 1438 |
+
"step": 10200
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.8704790819630365,
|
| 1442 |
+
"grad_norm": 126.49415588378906,
|
| 1443 |
+
"learning_rate": 6.595542501727713e-06,
|
| 1444 |
+
"loss": 19.93487548828125,
|
| 1445 |
+
"step": 10250
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.8747253213872465,
|
| 1449 |
+
"grad_norm": 130.64857482910156,
|
| 1450 |
+
"learning_rate": 6.379578438147893e-06,
|
| 1451 |
+
"loss": 19.3108447265625,
|
| 1452 |
+
"step": 10300
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.8789715608114563,
|
| 1456 |
+
"grad_norm": 155.81996154785156,
|
| 1457 |
+
"learning_rate": 6.1636143745680715e-06,
|
| 1458 |
+
"loss": 19.91413330078125,
|
| 1459 |
+
"step": 10350
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.8832178002356663,
|
| 1463 |
+
"grad_norm": 158.0801544189453,
|
| 1464 |
+
"learning_rate": 5.947650310988251e-06,
|
| 1465 |
+
"loss": 19.63139892578125,
|
| 1466 |
+
"step": 10400
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.8874640396598762,
|
| 1470 |
+
"grad_norm": 138.51763916015625,
|
| 1471 |
+
"learning_rate": 5.731686247408431e-06,
|
| 1472 |
+
"loss": 19.44971923828125,
|
| 1473 |
+
"step": 10450
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.8917102790840862,
|
| 1477 |
+
"grad_norm": 200.97584533691406,
|
| 1478 |
+
"learning_rate": 5.515722183828611e-06,
|
| 1479 |
+
"loss": 19.995697021484375,
|
| 1480 |
+
"step": 10500
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.895956518508296,
|
| 1484 |
+
"grad_norm": 125.70748901367188,
|
| 1485 |
+
"learning_rate": 5.299758120248791e-06,
|
| 1486 |
+
"loss": 19.891861572265626,
|
| 1487 |
+
"step": 10550
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.900202757932506,
|
| 1491 |
+
"grad_norm": 175.50962829589844,
|
| 1492 |
+
"learning_rate": 5.0837940566689705e-06,
|
| 1493 |
+
"loss": 19.681768798828124,
|
| 1494 |
+
"step": 10600
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.9044489973567159,
|
| 1498 |
+
"grad_norm": 129.02464294433594,
|
| 1499 |
+
"learning_rate": 4.86782999308915e-06,
|
| 1500 |
+
"loss": 19.740001220703125,
|
| 1501 |
+
"step": 10650
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.9086952367809259,
|
| 1505 |
+
"grad_norm": 170.73489379882812,
|
| 1506 |
+
"learning_rate": 4.651865929509329e-06,
|
| 1507 |
+
"loss": 20.2353564453125,
|
| 1508 |
+
"step": 10700
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.9129414762051358,
|
| 1512 |
+
"grad_norm": 141.15484619140625,
|
| 1513 |
+
"learning_rate": 4.435901865929509e-06,
|
| 1514 |
+
"loss": 19.27646240234375,
|
| 1515 |
+
"step": 10750
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.9171877156293458,
|
| 1519 |
+
"grad_norm": 139.16778564453125,
|
| 1520 |
+
"learning_rate": 4.219937802349689e-06,
|
| 1521 |
+
"loss": 20.000045166015624,
|
| 1522 |
+
"step": 10800
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.9214339550535557,
|
| 1526 |
+
"grad_norm": 160.3579559326172,
|
| 1527 |
+
"learning_rate": 4.003973738769869e-06,
|
| 1528 |
+
"loss": 18.942919921875,
|
| 1529 |
+
"step": 10850
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.9256801944777656,
|
| 1533 |
+
"grad_norm": 133.4417724609375,
|
| 1534 |
+
"learning_rate": 3.7880096751900484e-06,
|
| 1535 |
+
"loss": 19.30787841796875,
|
| 1536 |
+
"step": 10900
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.9299264339019756,
|
| 1540 |
+
"grad_norm": 147.41517639160156,
|
| 1541 |
+
"learning_rate": 3.572045611610228e-06,
|
| 1542 |
+
"loss": 19.907716064453126,
|
| 1543 |
+
"step": 10950
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.9341726733261855,
|
| 1547 |
+
"grad_norm": 137.2278289794922,
|
| 1548 |
+
"learning_rate": 3.3560815480304076e-06,
|
| 1549 |
+
"loss": 19.394835205078124,
|
| 1550 |
+
"step": 11000
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.9384189127503955,
|
| 1554 |
+
"grad_norm": 148.608642578125,
|
| 1555 |
+
"learning_rate": 3.1401174844505873e-06,
|
| 1556 |
+
"loss": 19.870865478515626,
|
| 1557 |
+
"step": 11050
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.9426651521746053,
|
| 1561 |
+
"grad_norm": 148.3636016845703,
|
| 1562 |
+
"learning_rate": 2.9241534208707676e-06,
|
| 1563 |
+
"loss": 20.17911376953125,
|
| 1564 |
+
"step": 11100
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.9469113915988153,
|
| 1568 |
+
"grad_norm": 136.2633056640625,
|
| 1569 |
+
"learning_rate": 2.708189357290947e-06,
|
| 1570 |
+
"loss": 19.6819775390625,
|
| 1571 |
+
"step": 11150
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.9511576310230252,
|
| 1575 |
+
"grad_norm": 132.42396545410156,
|
| 1576 |
+
"learning_rate": 2.4922252937111267e-06,
|
| 1577 |
+
"loss": 19.143297119140627,
|
| 1578 |
+
"step": 11200
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.9554038704472352,
|
| 1582 |
+
"grad_norm": 137.15524291992188,
|
| 1583 |
+
"learning_rate": 2.2762612301313065e-06,
|
| 1584 |
+
"loss": 19.832156982421875,
|
| 1585 |
+
"step": 11250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.9596501098714451,
|
| 1589 |
+
"grad_norm": 145.9365234375,
|
| 1590 |
+
"learning_rate": 2.060297166551486e-06,
|
| 1591 |
+
"loss": 19.251436767578124,
|
| 1592 |
+
"step": 11300
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.963896349295655,
|
| 1596 |
+
"grad_norm": 142.9067840576172,
|
| 1597 |
+
"learning_rate": 1.8443331029716657e-06,
|
| 1598 |
+
"loss": 19.578409423828123,
|
| 1599 |
+
"step": 11350
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.9681425887198649,
|
| 1603 |
+
"grad_norm": 160.01657104492188,
|
| 1604 |
+
"learning_rate": 1.6283690393918455e-06,
|
| 1605 |
+
"loss": 19.1957666015625,
|
| 1606 |
+
"step": 11400
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.9723888281440749,
|
| 1610 |
+
"grad_norm": 138.53872680664062,
|
| 1611 |
+
"learning_rate": 1.4124049758120248e-06,
|
| 1612 |
+
"loss": 18.92045654296875,
|
| 1613 |
+
"step": 11450
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.9766350675682849,
|
| 1617 |
+
"grad_norm": 150.38726806640625,
|
| 1618 |
+
"learning_rate": 1.1964409122322046e-06,
|
| 1619 |
+
"loss": 19.44462890625,
|
| 1620 |
+
"step": 11500
|
| 1621 |
+
}
|
| 1622 |
+
],
|
| 1623 |
+
"logging_steps": 50,
|
| 1624 |
+
"max_steps": 11776,
|
| 1625 |
+
"num_input_tokens_seen": 0,
|
| 1626 |
+
"num_train_epochs": 1,
|
| 1627 |
+
"save_steps": 500,
|
| 1628 |
+
"stateful_callbacks": {
|
| 1629 |
+
"TrainerControl": {
|
| 1630 |
+
"args": {
|
| 1631 |
+
"should_epoch_stop": false,
|
| 1632 |
+
"should_evaluate": false,
|
| 1633 |
+
"should_log": false,
|
| 1634 |
+
"should_save": true,
|
| 1635 |
+
"should_training_stop": false
|
| 1636 |
+
},
|
| 1637 |
+
"attributes": {}
|
| 1638 |
+
}
|
| 1639 |
+
},
|
| 1640 |
+
"total_flos": 3.3584807135739816e+16,
|
| 1641 |
+
"train_batch_size": 2,
|
| 1642 |
+
"trial_name": null,
|
| 1643 |
+
"trial_params": null
|
| 1644 |
+
}
|
checkpoint-11500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:38a7040df54471eb85e7c60e05f231ca54e9878768f7e376b60a2a2f85c48eae
|
| 3 |
+
size 5137
|
checkpoint-11776/config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_cross_attention": false,
|
| 3 |
+
"architectures": [
|
| 4 |
+
"XLMRobertaForMaskedLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"dtype": "float32",
|
| 10 |
+
"eos_token_id": 2,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.1,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"is_decoder": false,
|
| 17 |
+
"layer_norm_eps": 1e-05,
|
| 18 |
+
"max_position_embeddings": 514,
|
| 19 |
+
"model_type": "xlm-roberta",
|
| 20 |
+
"num_attention_heads": 12,
|
| 21 |
+
"num_hidden_layers": 12,
|
| 22 |
+
"output_past": true,
|
| 23 |
+
"pad_token_id": 1,
|
| 24 |
+
"position_embedding_type": "absolute",
|
| 25 |
+
"tie_word_embeddings": true,
|
| 26 |
+
"transformers_version": "5.10.2",
|
| 27 |
+
"type_vocab_size": 1,
|
| 28 |
+
"use_cache": false,
|
| 29 |
+
"vocab_size": 250002
|
| 30 |
+
}
|
checkpoint-11776/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d1d2be64d5ac753fa2bbeda1f7b419a3efb6365ca51e5ec6ab3c653872bfb8a1
|
| 3 |
+
size 1113205088
|
checkpoint-11776/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:87b78a1111ffd45c6b2346e804f36a9e62c9fe1edc23e44147cd206d5fda00d2
|
| 3 |
+
size 2226533707
|
checkpoint-11776/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b03ea3d43aff5498f5481ccbffd94c01dbd7b9d7e27c6b901d5744600c1d5259
|
| 3 |
+
size 14645
|
checkpoint-11776/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:94fa8dcf5b1e53d52564e9400ffdf8e0d624ebc8304c576de69142690ec206f5
|
| 3 |
+
size 1465
|
checkpoint-11776/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:acbd420e2269cdc1ef45332d3d5c418be4aef6b8cb5a0b7ccae0893485307153
|
| 3 |
+
size 17098086
|
checkpoint-11776/tokenizer_config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": true,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<s>",
|
| 5 |
+
"cls_token": "<s>",
|
| 6 |
+
"eos_token": "</s>",
|
| 7 |
+
"is_local": false,
|
| 8 |
+
"local_files_only": false,
|
| 9 |
+
"mask_token": "<mask>",
|
| 10 |
+
"model_max_length": 512,
|
| 11 |
+
"pad_token": "<pad>",
|
| 12 |
+
"sep_token": "</s>",
|
| 13 |
+
"tokenizer_class": "XLMRobertaTokenizer",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
checkpoint-11776/trainer_state.json
ADDED
|
@@ -0,0 +1,1679 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 11776,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.004246239424209934,
|
| 14 |
+
"grad_norm": 324.3861999511719,
|
| 15 |
+
"learning_rate": 1.225e-05,
|
| 16 |
+
"loss": 83.27494140625,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.008492478848419869,
|
| 21 |
+
"grad_norm": 294.6669921875,
|
| 22 |
+
"learning_rate": 2.4750000000000002e-05,
|
| 23 |
+
"loss": 72.928427734375,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.012738718272629802,
|
| 28 |
+
"grad_norm": 254.57310485839844,
|
| 29 |
+
"learning_rate": 3.7250000000000004e-05,
|
| 30 |
+
"loss": 65.434755859375,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.016984957696839737,
|
| 35 |
+
"grad_norm": 285.2427062988281,
|
| 36 |
+
"learning_rate": 4.975e-05,
|
| 37 |
+
"loss": 59.52314453125,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.02123119712104967,
|
| 42 |
+
"grad_norm": 276.11651611328125,
|
| 43 |
+
"learning_rate": 4.978835521769178e-05,
|
| 44 |
+
"loss": 56.26099609375,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.025477436545259604,
|
| 49 |
+
"grad_norm": 264.0826416015625,
|
| 50 |
+
"learning_rate": 4.9572391154111955e-05,
|
| 51 |
+
"loss": 52.364111328125,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.029723675969469537,
|
| 56 |
+
"grad_norm": 237.80140686035156,
|
| 57 |
+
"learning_rate": 4.9356427090532136e-05,
|
| 58 |
+
"loss": 49.438955078125,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.033969915393679474,
|
| 63 |
+
"grad_norm": 245.1451873779297,
|
| 64 |
+
"learning_rate": 4.914046302695232e-05,
|
| 65 |
+
"loss": 48.1695849609375,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.038216154817889404,
|
| 70 |
+
"grad_norm": 207.61068725585938,
|
| 71 |
+
"learning_rate": 4.892449896337249e-05,
|
| 72 |
+
"loss": 45.5499365234375,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.04246239424209934,
|
| 77 |
+
"grad_norm": 200.3720703125,
|
| 78 |
+
"learning_rate": 4.870853489979268e-05,
|
| 79 |
+
"loss": 44.5572998046875,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.04670863366630928,
|
| 84 |
+
"grad_norm": 197.68887329101562,
|
| 85 |
+
"learning_rate": 4.8492570836212855e-05,
|
| 86 |
+
"loss": 42.489599609375,
|
| 87 |
+
"step": 550
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.05095487309051921,
|
| 91 |
+
"grad_norm": 179.29957580566406,
|
| 92 |
+
"learning_rate": 4.827660677263304e-05,
|
| 93 |
+
"loss": 41.3450439453125,
|
| 94 |
+
"step": 600
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.055201112514729145,
|
| 98 |
+
"grad_norm": 226.55911254882812,
|
| 99 |
+
"learning_rate": 4.806064270905322e-05,
|
| 100 |
+
"loss": 41.2511376953125,
|
| 101 |
+
"step": 650
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.059447351938939075,
|
| 105 |
+
"grad_norm": 202.10791015625,
|
| 106 |
+
"learning_rate": 4.784467864547339e-05,
|
| 107 |
+
"loss": 39.70830322265625,
|
| 108 |
+
"step": 700
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.063693591363149,
|
| 112 |
+
"grad_norm": 207.3136444091797,
|
| 113 |
+
"learning_rate": 4.7628714581893575e-05,
|
| 114 |
+
"loss": 39.89173828125,
|
| 115 |
+
"step": 750
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.06793983078735895,
|
| 119 |
+
"grad_norm": 189.1178436279297,
|
| 120 |
+
"learning_rate": 4.7412750518313756e-05,
|
| 121 |
+
"loss": 38.6139501953125,
|
| 122 |
+
"step": 800
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.07218607021156888,
|
| 126 |
+
"grad_norm": 190.94015502929688,
|
| 127 |
+
"learning_rate": 4.719678645473393e-05,
|
| 128 |
+
"loss": 38.68489501953125,
|
| 129 |
+
"step": 850
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.07643230963577881,
|
| 133 |
+
"grad_norm": 191.75172424316406,
|
| 134 |
+
"learning_rate": 4.698082239115412e-05,
|
| 135 |
+
"loss": 36.644111328125,
|
| 136 |
+
"step": 900
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.08067854905998875,
|
| 140 |
+
"grad_norm": 206.9021453857422,
|
| 141 |
+
"learning_rate": 4.6764858327574294e-05,
|
| 142 |
+
"loss": 37.54802978515625,
|
| 143 |
+
"step": 950
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.08492478848419868,
|
| 147 |
+
"grad_norm": 184.31649780273438,
|
| 148 |
+
"learning_rate": 4.6548894263994475e-05,
|
| 149 |
+
"loss": 35.32251953125,
|
| 150 |
+
"step": 1000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.08917102790840861,
|
| 154 |
+
"grad_norm": 190.4438934326172,
|
| 155 |
+
"learning_rate": 4.633293020041466e-05,
|
| 156 |
+
"loss": 35.93584228515625,
|
| 157 |
+
"step": 1050
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.09341726733261856,
|
| 161 |
+
"grad_norm": 216.86279296875,
|
| 162 |
+
"learning_rate": 4.611696613683483e-05,
|
| 163 |
+
"loss": 34.81772216796875,
|
| 164 |
+
"step": 1100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.09766350675682849,
|
| 168 |
+
"grad_norm": 179.98419189453125,
|
| 169 |
+
"learning_rate": 4.590100207325501e-05,
|
| 170 |
+
"loss": 34.99229248046875,
|
| 171 |
+
"step": 1150
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.10190974618103842,
|
| 175 |
+
"grad_norm": 186.61732482910156,
|
| 176 |
+
"learning_rate": 4.568503800967519e-05,
|
| 177 |
+
"loss": 33.92795654296875,
|
| 178 |
+
"step": 1200
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.10615598560524835,
|
| 182 |
+
"grad_norm": 185.78001403808594,
|
| 183 |
+
"learning_rate": 4.546907394609537e-05,
|
| 184 |
+
"loss": 34.2297021484375,
|
| 185 |
+
"step": 1250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.11040222502945829,
|
| 189 |
+
"grad_norm": 236.7795867919922,
|
| 190 |
+
"learning_rate": 4.525310988251556e-05,
|
| 191 |
+
"loss": 33.76176513671875,
|
| 192 |
+
"step": 1300
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.11464846445366822,
|
| 196 |
+
"grad_norm": 210.53762817382812,
|
| 197 |
+
"learning_rate": 4.503714581893573e-05,
|
| 198 |
+
"loss": 32.79689453125,
|
| 199 |
+
"step": 1350
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.11889470387787815,
|
| 203 |
+
"grad_norm": 278.4184875488281,
|
| 204 |
+
"learning_rate": 4.4821181755355913e-05,
|
| 205 |
+
"loss": 31.72260009765625,
|
| 206 |
+
"step": 1400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.1231409433020881,
|
| 210 |
+
"grad_norm": 255.81129455566406,
|
| 211 |
+
"learning_rate": 4.460521769177609e-05,
|
| 212 |
+
"loss": 32.5959912109375,
|
| 213 |
+
"step": 1450
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.127387182726298,
|
| 217 |
+
"grad_norm": 188.4405975341797,
|
| 218 |
+
"learning_rate": 4.438925362819627e-05,
|
| 219 |
+
"loss": 31.72834716796875,
|
| 220 |
+
"step": 1500
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.13163342215050797,
|
| 224 |
+
"grad_norm": 196.35501098632812,
|
| 225 |
+
"learning_rate": 4.417328956461645e-05,
|
| 226 |
+
"loss": 32.69740478515625,
|
| 227 |
+
"step": 1550
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.1358796615747179,
|
| 231 |
+
"grad_norm": 186.20811462402344,
|
| 232 |
+
"learning_rate": 4.3957325501036626e-05,
|
| 233 |
+
"loss": 31.13796875,
|
| 234 |
+
"step": 1600
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.14012590099892783,
|
| 238 |
+
"grad_norm": 158.75543212890625,
|
| 239 |
+
"learning_rate": 4.374136143745681e-05,
|
| 240 |
+
"loss": 32.08728271484375,
|
| 241 |
+
"step": 1650
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.14437214042313776,
|
| 245 |
+
"grad_norm": 207.9669952392578,
|
| 246 |
+
"learning_rate": 4.352539737387699e-05,
|
| 247 |
+
"loss": 31.42122802734375,
|
| 248 |
+
"step": 1700
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.1486183798473477,
|
| 252 |
+
"grad_norm": 184.7431640625,
|
| 253 |
+
"learning_rate": 4.330943331029717e-05,
|
| 254 |
+
"loss": 31.1582763671875,
|
| 255 |
+
"step": 1750
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.15286461927155762,
|
| 259 |
+
"grad_norm": 155.47659301757812,
|
| 260 |
+
"learning_rate": 4.309346924671735e-05,
|
| 261 |
+
"loss": 30.77266845703125,
|
| 262 |
+
"step": 1800
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.15711085869576757,
|
| 266 |
+
"grad_norm": 163.03013610839844,
|
| 267 |
+
"learning_rate": 4.2877505183137527e-05,
|
| 268 |
+
"loss": 30.72901123046875,
|
| 269 |
+
"step": 1850
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.1613570981199775,
|
| 273 |
+
"grad_norm": 240.4451904296875,
|
| 274 |
+
"learning_rate": 4.266154111955771e-05,
|
| 275 |
+
"loss": 30.06856689453125,
|
| 276 |
+
"step": 1900
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.16560333754418743,
|
| 280 |
+
"grad_norm": 193.2200164794922,
|
| 281 |
+
"learning_rate": 4.244557705597789e-05,
|
| 282 |
+
"loss": 29.64366455078125,
|
| 283 |
+
"step": 1950
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.16984957696839736,
|
| 287 |
+
"grad_norm": 197.7560577392578,
|
| 288 |
+
"learning_rate": 4.2229612992398064e-05,
|
| 289 |
+
"loss": 29.403779296875,
|
| 290 |
+
"step": 2000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.1740958163926073,
|
| 294 |
+
"grad_norm": 172.2969512939453,
|
| 295 |
+
"learning_rate": 4.2013648928818246e-05,
|
| 296 |
+
"loss": 29.76959716796875,
|
| 297 |
+
"step": 2050
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.17834205581681722,
|
| 301 |
+
"grad_norm": 171.65423583984375,
|
| 302 |
+
"learning_rate": 4.179768486523843e-05,
|
| 303 |
+
"loss": 28.84666259765625,
|
| 304 |
+
"step": 2100
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.18258829524102715,
|
| 308 |
+
"grad_norm": 253.046875,
|
| 309 |
+
"learning_rate": 4.158172080165861e-05,
|
| 310 |
+
"loss": 28.4874462890625,
|
| 311 |
+
"step": 2150
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.1868345346652371,
|
| 315 |
+
"grad_norm": 174.5941162109375,
|
| 316 |
+
"learning_rate": 4.136575673807879e-05,
|
| 317 |
+
"loss": 28.8812744140625,
|
| 318 |
+
"step": 2200
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.19108077408944704,
|
| 322 |
+
"grad_norm": 187.45791625976562,
|
| 323 |
+
"learning_rate": 4.1149792674498965e-05,
|
| 324 |
+
"loss": 29.3225390625,
|
| 325 |
+
"step": 2250
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.19532701351365697,
|
| 329 |
+
"grad_norm": 158.6984405517578,
|
| 330 |
+
"learning_rate": 4.0933828610919146e-05,
|
| 331 |
+
"loss": 29.0456494140625,
|
| 332 |
+
"step": 2300
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.1995732529378669,
|
| 336 |
+
"grad_norm": 162.81512451171875,
|
| 337 |
+
"learning_rate": 4.071786454733932e-05,
|
| 338 |
+
"loss": 29.16901123046875,
|
| 339 |
+
"step": 2350
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.20381949236207683,
|
| 343 |
+
"grad_norm": 201.16639709472656,
|
| 344 |
+
"learning_rate": 4.05019004837595e-05,
|
| 345 |
+
"loss": 29.002841796875,
|
| 346 |
+
"step": 2400
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.20806573178628676,
|
| 350 |
+
"grad_norm": 185.6593475341797,
|
| 351 |
+
"learning_rate": 4.0285936420179684e-05,
|
| 352 |
+
"loss": 28.537255859375,
|
| 353 |
+
"step": 2450
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.2123119712104967,
|
| 357 |
+
"grad_norm": 166.38766479492188,
|
| 358 |
+
"learning_rate": 4.0069972356599866e-05,
|
| 359 |
+
"loss": 27.69727294921875,
|
| 360 |
+
"step": 2500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.21655821063470665,
|
| 364 |
+
"grad_norm": 174.38104248046875,
|
| 365 |
+
"learning_rate": 3.985400829302005e-05,
|
| 366 |
+
"loss": 28.5172802734375,
|
| 367 |
+
"step": 2550
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.22080445005891658,
|
| 371 |
+
"grad_norm": 149.02406311035156,
|
| 372 |
+
"learning_rate": 3.963804422944022e-05,
|
| 373 |
+
"loss": 28.4050244140625,
|
| 374 |
+
"step": 2600
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.2250506894831265,
|
| 378 |
+
"grad_norm": 192.12002563476562,
|
| 379 |
+
"learning_rate": 3.94220801658604e-05,
|
| 380 |
+
"loss": 26.98137451171875,
|
| 381 |
+
"step": 2650
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.22929692890733644,
|
| 385 |
+
"grad_norm": 170.00814819335938,
|
| 386 |
+
"learning_rate": 3.9206116102280585e-05,
|
| 387 |
+
"loss": 26.99474853515625,
|
| 388 |
+
"step": 2700
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.23354316833154637,
|
| 392 |
+
"grad_norm": 145.735595703125,
|
| 393 |
+
"learning_rate": 3.899015203870076e-05,
|
| 394 |
+
"loss": 27.34177001953125,
|
| 395 |
+
"step": 2750
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.2377894077557563,
|
| 399 |
+
"grad_norm": 169.86480712890625,
|
| 400 |
+
"learning_rate": 3.877418797512094e-05,
|
| 401 |
+
"loss": 27.52475830078125,
|
| 402 |
+
"step": 2800
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.24203564717996623,
|
| 406 |
+
"grad_norm": 135.18838500976562,
|
| 407 |
+
"learning_rate": 3.855822391154112e-05,
|
| 408 |
+
"loss": 27.2542626953125,
|
| 409 |
+
"step": 2850
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.2462818866041762,
|
| 413 |
+
"grad_norm": 149.96510314941406,
|
| 414 |
+
"learning_rate": 3.8342259847961304e-05,
|
| 415 |
+
"loss": 27.07297119140625,
|
| 416 |
+
"step": 2900
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.2505281260283861,
|
| 420 |
+
"grad_norm": 159.20584106445312,
|
| 421 |
+
"learning_rate": 3.8126295784381485e-05,
|
| 422 |
+
"loss": 27.08953125,
|
| 423 |
+
"step": 2950
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.254774365452596,
|
| 427 |
+
"grad_norm": 136.2477569580078,
|
| 428 |
+
"learning_rate": 3.791033172080166e-05,
|
| 429 |
+
"loss": 26.437490234375,
|
| 430 |
+
"step": 3000
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.259020604876806,
|
| 434 |
+
"grad_norm": 164.90737915039062,
|
| 435 |
+
"learning_rate": 3.769436765722184e-05,
|
| 436 |
+
"loss": 26.1881103515625,
|
| 437 |
+
"step": 3050
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.26326684430101593,
|
| 441 |
+
"grad_norm": 154.27102661132812,
|
| 442 |
+
"learning_rate": 3.747840359364202e-05,
|
| 443 |
+
"loss": 26.475185546875,
|
| 444 |
+
"step": 3100
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.26751308372522586,
|
| 448 |
+
"grad_norm": 174.78500366210938,
|
| 449 |
+
"learning_rate": 3.72624395300622e-05,
|
| 450 |
+
"loss": 26.410517578125,
|
| 451 |
+
"step": 3150
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.2717593231494358,
|
| 455 |
+
"grad_norm": 167.15457153320312,
|
| 456 |
+
"learning_rate": 3.704647546648238e-05,
|
| 457 |
+
"loss": 25.28077880859375,
|
| 458 |
+
"step": 3200
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.2760055625736457,
|
| 462 |
+
"grad_norm": 182.3014373779297,
|
| 463 |
+
"learning_rate": 3.6830511402902554e-05,
|
| 464 |
+
"loss": 26.1413818359375,
|
| 465 |
+
"step": 3250
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.28025180199785565,
|
| 469 |
+
"grad_norm": 185.13711547851562,
|
| 470 |
+
"learning_rate": 3.661454733932274e-05,
|
| 471 |
+
"loss": 25.593623046875,
|
| 472 |
+
"step": 3300
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.2844980414220656,
|
| 476 |
+
"grad_norm": 136.2028045654297,
|
| 477 |
+
"learning_rate": 3.639858327574292e-05,
|
| 478 |
+
"loss": 26.10810791015625,
|
| 479 |
+
"step": 3350
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.2887442808462755,
|
| 483 |
+
"grad_norm": 160.10540771484375,
|
| 484 |
+
"learning_rate": 3.61826192121631e-05,
|
| 485 |
+
"loss": 26.09581298828125,
|
| 486 |
+
"step": 3400
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.29299052027048544,
|
| 490 |
+
"grad_norm": 170.12704467773438,
|
| 491 |
+
"learning_rate": 3.596665514858328e-05,
|
| 492 |
+
"loss": 26.3189306640625,
|
| 493 |
+
"step": 3450
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.2972367596946954,
|
| 497 |
+
"grad_norm": 154.6916961669922,
|
| 498 |
+
"learning_rate": 3.5750691085003455e-05,
|
| 499 |
+
"loss": 25.707509765625,
|
| 500 |
+
"step": 3500
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.3014829991189053,
|
| 504 |
+
"grad_norm": 175.77297973632812,
|
| 505 |
+
"learning_rate": 3.5534727021423636e-05,
|
| 506 |
+
"loss": 26.1056396484375,
|
| 507 |
+
"step": 3550
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.30572923854311523,
|
| 511 |
+
"grad_norm": 143.42288208007812,
|
| 512 |
+
"learning_rate": 3.531876295784382e-05,
|
| 513 |
+
"loss": 25.2949560546875,
|
| 514 |
+
"step": 3600
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.30997547796732516,
|
| 518 |
+
"grad_norm": 171.7968292236328,
|
| 519 |
+
"learning_rate": 3.510279889426399e-05,
|
| 520 |
+
"loss": 25.76105712890625,
|
| 521 |
+
"step": 3650
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.31422171739153515,
|
| 525 |
+
"grad_norm": 165.98411560058594,
|
| 526 |
+
"learning_rate": 3.488683483068418e-05,
|
| 527 |
+
"loss": 25.60622802734375,
|
| 528 |
+
"step": 3700
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.3184679568157451,
|
| 532 |
+
"grad_norm": 157.76429748535156,
|
| 533 |
+
"learning_rate": 3.4670870767104355e-05,
|
| 534 |
+
"loss": 24.7918017578125,
|
| 535 |
+
"step": 3750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.322714196239955,
|
| 539 |
+
"grad_norm": 168.3741455078125,
|
| 540 |
+
"learning_rate": 3.445490670352454e-05,
|
| 541 |
+
"loss": 24.83272705078125,
|
| 542 |
+
"step": 3800
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.32696043566416494,
|
| 546 |
+
"grad_norm": 165.90496826171875,
|
| 547 |
+
"learning_rate": 3.423894263994472e-05,
|
| 548 |
+
"loss": 26.2589599609375,
|
| 549 |
+
"step": 3850
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.33120667508837487,
|
| 553 |
+
"grad_norm": 142.2030792236328,
|
| 554 |
+
"learning_rate": 3.402297857636489e-05,
|
| 555 |
+
"loss": 24.28029541015625,
|
| 556 |
+
"step": 3900
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.3354529145125848,
|
| 560 |
+
"grad_norm": 159.05149841308594,
|
| 561 |
+
"learning_rate": 3.3807014512785074e-05,
|
| 562 |
+
"loss": 24.80717041015625,
|
| 563 |
+
"step": 3950
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.33969915393679473,
|
| 567 |
+
"grad_norm": 138.15611267089844,
|
| 568 |
+
"learning_rate": 3.359105044920525e-05,
|
| 569 |
+
"loss": 24.88008056640625,
|
| 570 |
+
"step": 4000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.34394539336100466,
|
| 574 |
+
"grad_norm": 196.62498474121094,
|
| 575 |
+
"learning_rate": 3.337508638562543e-05,
|
| 576 |
+
"loss": 24.37576171875,
|
| 577 |
+
"step": 4050
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.3481916327852146,
|
| 581 |
+
"grad_norm": 155.42593383789062,
|
| 582 |
+
"learning_rate": 3.315912232204562e-05,
|
| 583 |
+
"loss": 25.1333447265625,
|
| 584 |
+
"step": 4100
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.3524378722094245,
|
| 588 |
+
"grad_norm": 146.24502563476562,
|
| 589 |
+
"learning_rate": 3.2943158258465794e-05,
|
| 590 |
+
"loss": 23.95904296875,
|
| 591 |
+
"step": 4150
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.35668411163363445,
|
| 595 |
+
"grad_norm": 140.36927795410156,
|
| 596 |
+
"learning_rate": 3.2727194194885975e-05,
|
| 597 |
+
"loss": 24.21125,
|
| 598 |
+
"step": 4200
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.3609303510578444,
|
| 602 |
+
"grad_norm": 151.30316162109375,
|
| 603 |
+
"learning_rate": 3.251123013130615e-05,
|
| 604 |
+
"loss": 24.09343505859375,
|
| 605 |
+
"step": 4250
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.3651765904820543,
|
| 609 |
+
"grad_norm": 209.92578125,
|
| 610 |
+
"learning_rate": 3.229526606772633e-05,
|
| 611 |
+
"loss": 23.6073779296875,
|
| 612 |
+
"step": 4300
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.36942282990626424,
|
| 616 |
+
"grad_norm": 162.1195526123047,
|
| 617 |
+
"learning_rate": 3.207930200414651e-05,
|
| 618 |
+
"loss": 24.49541748046875,
|
| 619 |
+
"step": 4350
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.3736690693304742,
|
| 623 |
+
"grad_norm": 170.38694763183594,
|
| 624 |
+
"learning_rate": 3.186333794056669e-05,
|
| 625 |
+
"loss": 24.1321630859375,
|
| 626 |
+
"step": 4400
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.37791530875468415,
|
| 630 |
+
"grad_norm": 161.4326934814453,
|
| 631 |
+
"learning_rate": 3.164737387698687e-05,
|
| 632 |
+
"loss": 23.7575341796875,
|
| 633 |
+
"step": 4450
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.3821615481788941,
|
| 637 |
+
"grad_norm": 171.4528045654297,
|
| 638 |
+
"learning_rate": 3.143140981340705e-05,
|
| 639 |
+
"loss": 24.50120849609375,
|
| 640 |
+
"step": 4500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.386407787603104,
|
| 644 |
+
"grad_norm": 160.71112060546875,
|
| 645 |
+
"learning_rate": 3.121544574982723e-05,
|
| 646 |
+
"loss": 23.270244140625,
|
| 647 |
+
"step": 4550
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.39065402702731394,
|
| 651 |
+
"grad_norm": 139.88951110839844,
|
| 652 |
+
"learning_rate": 3.099948168624741e-05,
|
| 653 |
+
"loss": 24.25181640625,
|
| 654 |
+
"step": 4600
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.3949002664515239,
|
| 658 |
+
"grad_norm": 152.9877166748047,
|
| 659 |
+
"learning_rate": 3.078351762266759e-05,
|
| 660 |
+
"loss": 23.2979345703125,
|
| 661 |
+
"step": 4650
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.3991465058757338,
|
| 665 |
+
"grad_norm": 179.30581665039062,
|
| 666 |
+
"learning_rate": 3.056755355908777e-05,
|
| 667 |
+
"loss": 24.148583984375,
|
| 668 |
+
"step": 4700
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.40339274529994373,
|
| 672 |
+
"grad_norm": 143.60121154785156,
|
| 673 |
+
"learning_rate": 3.0351589495507948e-05,
|
| 674 |
+
"loss": 24.34101318359375,
|
| 675 |
+
"step": 4750
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.40763898472415366,
|
| 679 |
+
"grad_norm": 133.65771484375,
|
| 680 |
+
"learning_rate": 3.0135625431928126e-05,
|
| 681 |
+
"loss": 23.299140625,
|
| 682 |
+
"step": 4800
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.4118852241483636,
|
| 686 |
+
"grad_norm": 162.5352783203125,
|
| 687 |
+
"learning_rate": 2.9919661368348307e-05,
|
| 688 |
+
"loss": 23.08830322265625,
|
| 689 |
+
"step": 4850
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.4161314635725735,
|
| 693 |
+
"grad_norm": 158.4131317138672,
|
| 694 |
+
"learning_rate": 2.9703697304768485e-05,
|
| 695 |
+
"loss": 22.8610205078125,
|
| 696 |
+
"step": 4900
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.42037770299678345,
|
| 700 |
+
"grad_norm": 124.68849182128906,
|
| 701 |
+
"learning_rate": 2.948773324118867e-05,
|
| 702 |
+
"loss": 23.79708740234375,
|
| 703 |
+
"step": 4950
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.4246239424209934,
|
| 707 |
+
"grad_norm": 163.76657104492188,
|
| 708 |
+
"learning_rate": 2.9271769177608848e-05,
|
| 709 |
+
"loss": 23.52044189453125,
|
| 710 |
+
"step": 5000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.42887018184520337,
|
| 714 |
+
"grad_norm": 187.97413635253906,
|
| 715 |
+
"learning_rate": 2.9055805114029026e-05,
|
| 716 |
+
"loss": 24.02717529296875,
|
| 717 |
+
"step": 5050
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.4331164212694133,
|
| 721 |
+
"grad_norm": 154.0001220703125,
|
| 722 |
+
"learning_rate": 2.8839841050449208e-05,
|
| 723 |
+
"loss": 23.41170166015625,
|
| 724 |
+
"step": 5100
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.43736266069362323,
|
| 728 |
+
"grad_norm": 141.18080139160156,
|
| 729 |
+
"learning_rate": 2.8623876986869386e-05,
|
| 730 |
+
"loss": 23.554814453125,
|
| 731 |
+
"step": 5150
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.44160890011783316,
|
| 735 |
+
"grad_norm": 158.98216247558594,
|
| 736 |
+
"learning_rate": 2.8407912923289564e-05,
|
| 737 |
+
"loss": 23.31008544921875,
|
| 738 |
+
"step": 5200
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.4458551395420431,
|
| 742 |
+
"grad_norm": 146.218994140625,
|
| 743 |
+
"learning_rate": 2.8191948859709742e-05,
|
| 744 |
+
"loss": 23.71553955078125,
|
| 745 |
+
"step": 5250
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.450101378966253,
|
| 749 |
+
"grad_norm": 156.51646423339844,
|
| 750 |
+
"learning_rate": 2.7975984796129924e-05,
|
| 751 |
+
"loss": 22.9737060546875,
|
| 752 |
+
"step": 5300
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.45434761839046295,
|
| 756 |
+
"grad_norm": 144.48443603515625,
|
| 757 |
+
"learning_rate": 2.776002073255011e-05,
|
| 758 |
+
"loss": 24.7643798828125,
|
| 759 |
+
"step": 5350
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.4585938578146729,
|
| 763 |
+
"grad_norm": 160.95411682128906,
|
| 764 |
+
"learning_rate": 2.7544056668970287e-05,
|
| 765 |
+
"loss": 23.05649658203125,
|
| 766 |
+
"step": 5400
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.4628400972388828,
|
| 770 |
+
"grad_norm": 152.6392059326172,
|
| 771 |
+
"learning_rate": 2.7328092605390465e-05,
|
| 772 |
+
"loss": 23.15297119140625,
|
| 773 |
+
"step": 5450
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.46708633666309274,
|
| 777 |
+
"grad_norm": 190.5279083251953,
|
| 778 |
+
"learning_rate": 2.7112128541810643e-05,
|
| 779 |
+
"loss": 21.804638671875,
|
| 780 |
+
"step": 5500
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.47133257608730267,
|
| 784 |
+
"grad_norm": 170.51712036132812,
|
| 785 |
+
"learning_rate": 2.6896164478230824e-05,
|
| 786 |
+
"loss": 22.71973388671875,
|
| 787 |
+
"step": 5550
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.4755788155115126,
|
| 791 |
+
"grad_norm": 151.53326416015625,
|
| 792 |
+
"learning_rate": 2.6680200414651002e-05,
|
| 793 |
+
"loss": 22.90906494140625,
|
| 794 |
+
"step": 5600
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.47982505493572253,
|
| 798 |
+
"grad_norm": 147.06654357910156,
|
| 799 |
+
"learning_rate": 2.646423635107118e-05,
|
| 800 |
+
"loss": 24.11918701171875,
|
| 801 |
+
"step": 5650
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.48407129435993246,
|
| 805 |
+
"grad_norm": 167.34326171875,
|
| 806 |
+
"learning_rate": 2.624827228749136e-05,
|
| 807 |
+
"loss": 22.7313134765625,
|
| 808 |
+
"step": 5700
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.48831753378414244,
|
| 812 |
+
"grad_norm": 136.9669647216797,
|
| 813 |
+
"learning_rate": 2.6032308223911543e-05,
|
| 814 |
+
"loss": 21.70623046875,
|
| 815 |
+
"step": 5750
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.4925637732083524,
|
| 819 |
+
"grad_norm": 159.77626037597656,
|
| 820 |
+
"learning_rate": 2.5816344160331725e-05,
|
| 821 |
+
"loss": 22.6339453125,
|
| 822 |
+
"step": 5800
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.4968100126325623,
|
| 826 |
+
"grad_norm": 198.8340301513672,
|
| 827 |
+
"learning_rate": 2.5600380096751903e-05,
|
| 828 |
+
"loss": 22.50951171875,
|
| 829 |
+
"step": 5850
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.5010562520567722,
|
| 833 |
+
"grad_norm": 139.2235870361328,
|
| 834 |
+
"learning_rate": 2.538441603317208e-05,
|
| 835 |
+
"loss": 22.05862548828125,
|
| 836 |
+
"step": 5900
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.5053024914809822,
|
| 840 |
+
"grad_norm": 145.6766815185547,
|
| 841 |
+
"learning_rate": 2.516845196959226e-05,
|
| 842 |
+
"loss": 23.03335205078125,
|
| 843 |
+
"step": 5950
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.509548730905192,
|
| 847 |
+
"grad_norm": 159.59835815429688,
|
| 848 |
+
"learning_rate": 2.495248790601244e-05,
|
| 849 |
+
"loss": 23.1497705078125,
|
| 850 |
+
"step": 6000
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.513794970329402,
|
| 854 |
+
"grad_norm": 200.4053497314453,
|
| 855 |
+
"learning_rate": 2.4736523842432622e-05,
|
| 856 |
+
"loss": 22.751318359375,
|
| 857 |
+
"step": 6050
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.518041209753612,
|
| 861 |
+
"grad_norm": 146.4293670654297,
|
| 862 |
+
"learning_rate": 2.45205597788528e-05,
|
| 863 |
+
"loss": 22.441064453125,
|
| 864 |
+
"step": 6100
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.5222874491778219,
|
| 868 |
+
"grad_norm": 140.7433319091797,
|
| 869 |
+
"learning_rate": 2.430459571527298e-05,
|
| 870 |
+
"loss": 21.8677880859375,
|
| 871 |
+
"step": 6150
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.5265336886020319,
|
| 875 |
+
"grad_norm": 145.3983154296875,
|
| 876 |
+
"learning_rate": 2.4088631651693156e-05,
|
| 877 |
+
"loss": 21.99537109375,
|
| 878 |
+
"step": 6200
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.5307799280262417,
|
| 882 |
+
"grad_norm": 171.57505798339844,
|
| 883 |
+
"learning_rate": 2.387266758811334e-05,
|
| 884 |
+
"loss": 22.0473974609375,
|
| 885 |
+
"step": 6250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.5350261674504517,
|
| 889 |
+
"grad_norm": 190.87144470214844,
|
| 890 |
+
"learning_rate": 2.365670352453352e-05,
|
| 891 |
+
"loss": 22.20819580078125,
|
| 892 |
+
"step": 6300
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.5392724068746616,
|
| 896 |
+
"grad_norm": 131.15647888183594,
|
| 897 |
+
"learning_rate": 2.3440739460953698e-05,
|
| 898 |
+
"loss": 22.3381689453125,
|
| 899 |
+
"step": 6350
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.5435186462988716,
|
| 903 |
+
"grad_norm": 130.159423828125,
|
| 904 |
+
"learning_rate": 2.3224775397373876e-05,
|
| 905 |
+
"loss": 21.13576171875,
|
| 906 |
+
"step": 6400
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.5477648857230815,
|
| 910 |
+
"grad_norm": 153.32504272460938,
|
| 911 |
+
"learning_rate": 2.3008811333794057e-05,
|
| 912 |
+
"loss": 21.22865478515625,
|
| 913 |
+
"step": 6450
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.5520111251472914,
|
| 917 |
+
"grad_norm": 137.8641357421875,
|
| 918 |
+
"learning_rate": 2.279284727021424e-05,
|
| 919 |
+
"loss": 21.42198974609375,
|
| 920 |
+
"step": 6500
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.5562573645715013,
|
| 924 |
+
"grad_norm": 152.36447143554688,
|
| 925 |
+
"learning_rate": 2.2576883206634417e-05,
|
| 926 |
+
"loss": 22.49668212890625,
|
| 927 |
+
"step": 6550
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.5605036039957113,
|
| 931 |
+
"grad_norm": 170.3688507080078,
|
| 932 |
+
"learning_rate": 2.2360919143054595e-05,
|
| 933 |
+
"loss": 21.45054443359375,
|
| 934 |
+
"step": 6600
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.5647498434199212,
|
| 938 |
+
"grad_norm": 119.01740264892578,
|
| 939 |
+
"learning_rate": 2.2144955079474776e-05,
|
| 940 |
+
"loss": 22.2169287109375,
|
| 941 |
+
"step": 6650
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.5689960828441312,
|
| 945 |
+
"grad_norm": 147.85865783691406,
|
| 946 |
+
"learning_rate": 2.1928991015894958e-05,
|
| 947 |
+
"loss": 22.83664794921875,
|
| 948 |
+
"step": 6700
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.5732423222683412,
|
| 952 |
+
"grad_norm": 153.3690948486328,
|
| 953 |
+
"learning_rate": 2.1713026952315136e-05,
|
| 954 |
+
"loss": 21.9330712890625,
|
| 955 |
+
"step": 6750
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.577488561692551,
|
| 959 |
+
"grad_norm": 137.37164306640625,
|
| 960 |
+
"learning_rate": 2.1497062888735314e-05,
|
| 961 |
+
"loss": 20.79832763671875,
|
| 962 |
+
"step": 6800
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.581734801116761,
|
| 966 |
+
"grad_norm": 130.9550018310547,
|
| 967 |
+
"learning_rate": 2.1281098825155495e-05,
|
| 968 |
+
"loss": 21.09296630859375,
|
| 969 |
+
"step": 6850
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.5859810405409709,
|
| 973 |
+
"grad_norm": 121.02996063232422,
|
| 974 |
+
"learning_rate": 2.1065134761575674e-05,
|
| 975 |
+
"loss": 21.2236083984375,
|
| 976 |
+
"step": 6900
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.5902272799651809,
|
| 980 |
+
"grad_norm": 169.94479370117188,
|
| 981 |
+
"learning_rate": 2.0849170697995855e-05,
|
| 982 |
+
"loss": 21.0869140625,
|
| 983 |
+
"step": 6950
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.5944735193893907,
|
| 987 |
+
"grad_norm": 139.7215576171875,
|
| 988 |
+
"learning_rate": 2.0633206634416033e-05,
|
| 989 |
+
"loss": 20.95746337890625,
|
| 990 |
+
"step": 7000
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.5987197588136007,
|
| 994 |
+
"grad_norm": 135.5031280517578,
|
| 995 |
+
"learning_rate": 2.0417242570836215e-05,
|
| 996 |
+
"loss": 21.83375,
|
| 997 |
+
"step": 7050
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.6029659982378106,
|
| 1001 |
+
"grad_norm": 143.67396545410156,
|
| 1002 |
+
"learning_rate": 2.0201278507256393e-05,
|
| 1003 |
+
"loss": 21.9516015625,
|
| 1004 |
+
"step": 7100
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.6072122376620206,
|
| 1008 |
+
"grad_norm": 242.10548400878906,
|
| 1009 |
+
"learning_rate": 1.9985314443676574e-05,
|
| 1010 |
+
"loss": 21.42175537109375,
|
| 1011 |
+
"step": 7150
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.6114584770862305,
|
| 1015 |
+
"grad_norm": 157.11415100097656,
|
| 1016 |
+
"learning_rate": 1.9769350380096752e-05,
|
| 1017 |
+
"loss": 21.08699951171875,
|
| 1018 |
+
"step": 7200
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.6157047165104405,
|
| 1022 |
+
"grad_norm": 124.51484680175781,
|
| 1023 |
+
"learning_rate": 1.9553386316516934e-05,
|
| 1024 |
+
"loss": 21.26003662109375,
|
| 1025 |
+
"step": 7250
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.6199509559346503,
|
| 1029 |
+
"grad_norm": 175.8770751953125,
|
| 1030 |
+
"learning_rate": 1.9337422252937112e-05,
|
| 1031 |
+
"loss": 21.37485107421875,
|
| 1032 |
+
"step": 7300
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.6241971953588603,
|
| 1036 |
+
"grad_norm": 139.2329559326172,
|
| 1037 |
+
"learning_rate": 1.912145818935729e-05,
|
| 1038 |
+
"loss": 21.006982421875,
|
| 1039 |
+
"step": 7350
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.6284434347830703,
|
| 1043 |
+
"grad_norm": 142.4539031982422,
|
| 1044 |
+
"learning_rate": 1.890549412577747e-05,
|
| 1045 |
+
"loss": 20.92181640625,
|
| 1046 |
+
"step": 7400
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.6326896742072802,
|
| 1050 |
+
"grad_norm": 143.5320281982422,
|
| 1051 |
+
"learning_rate": 1.8689530062197653e-05,
|
| 1052 |
+
"loss": 21.85162109375,
|
| 1053 |
+
"step": 7450
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.6369359136314902,
|
| 1057 |
+
"grad_norm": 152.91880798339844,
|
| 1058 |
+
"learning_rate": 1.847356599861783e-05,
|
| 1059 |
+
"loss": 20.92374755859375,
|
| 1060 |
+
"step": 7500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.6411821530557,
|
| 1064 |
+
"grad_norm": 162.99484252929688,
|
| 1065 |
+
"learning_rate": 1.825760193503801e-05,
|
| 1066 |
+
"loss": 21.4075537109375,
|
| 1067 |
+
"step": 7550
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.64542839247991,
|
| 1071 |
+
"grad_norm": 164.03475952148438,
|
| 1072 |
+
"learning_rate": 1.804163787145819e-05,
|
| 1073 |
+
"loss": 21.6792919921875,
|
| 1074 |
+
"step": 7600
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.6496746319041199,
|
| 1078 |
+
"grad_norm": 154.84764099121094,
|
| 1079 |
+
"learning_rate": 1.7825673807878372e-05,
|
| 1080 |
+
"loss": 20.593271484375,
|
| 1081 |
+
"step": 7650
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.6539208713283299,
|
| 1085 |
+
"grad_norm": 136.49598693847656,
|
| 1086 |
+
"learning_rate": 1.760970974429855e-05,
|
| 1087 |
+
"loss": 20.4891650390625,
|
| 1088 |
+
"step": 7700
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.6581671107525398,
|
| 1092 |
+
"grad_norm": 144.41871643066406,
|
| 1093 |
+
"learning_rate": 1.739374568071873e-05,
|
| 1094 |
+
"loss": 21.7032177734375,
|
| 1095 |
+
"step": 7750
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.6624133501767497,
|
| 1099 |
+
"grad_norm": 172.35324096679688,
|
| 1100 |
+
"learning_rate": 1.7177781617138906e-05,
|
| 1101 |
+
"loss": 21.29265380859375,
|
| 1102 |
+
"step": 7800
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.6666595896009596,
|
| 1106 |
+
"grad_norm": 154.5519256591797,
|
| 1107 |
+
"learning_rate": 1.6961817553559088e-05,
|
| 1108 |
+
"loss": 20.79191650390625,
|
| 1109 |
+
"step": 7850
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.6709058290251696,
|
| 1113 |
+
"grad_norm": 157.8460235595703,
|
| 1114 |
+
"learning_rate": 1.674585348997927e-05,
|
| 1115 |
+
"loss": 20.80486328125,
|
| 1116 |
+
"step": 7900
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.6751520684493795,
|
| 1120 |
+
"grad_norm": 160.06161499023438,
|
| 1121 |
+
"learning_rate": 1.6529889426399447e-05,
|
| 1122 |
+
"loss": 20.06711669921875,
|
| 1123 |
+
"step": 7950
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.6793983078735895,
|
| 1127 |
+
"grad_norm": 150.2727813720703,
|
| 1128 |
+
"learning_rate": 1.6313925362819626e-05,
|
| 1129 |
+
"loss": 20.79498046875,
|
| 1130 |
+
"step": 8000
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.6836445472977993,
|
| 1134 |
+
"grad_norm": 144.0926971435547,
|
| 1135 |
+
"learning_rate": 1.6097961299239807e-05,
|
| 1136 |
+
"loss": 21.199130859375,
|
| 1137 |
+
"step": 8050
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.6878907867220093,
|
| 1141 |
+
"grad_norm": 149.16781616210938,
|
| 1142 |
+
"learning_rate": 1.588199723565999e-05,
|
| 1143 |
+
"loss": 20.359921875,
|
| 1144 |
+
"step": 8100
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.6921370261462193,
|
| 1148 |
+
"grad_norm": 167.05862426757812,
|
| 1149 |
+
"learning_rate": 1.5666033172080167e-05,
|
| 1150 |
+
"loss": 20.74350341796875,
|
| 1151 |
+
"step": 8150
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.6963832655704292,
|
| 1155 |
+
"grad_norm": 164.15008544921875,
|
| 1156 |
+
"learning_rate": 1.5450069108500345e-05,
|
| 1157 |
+
"loss": 19.6603076171875,
|
| 1158 |
+
"step": 8200
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.7006295049946392,
|
| 1162 |
+
"grad_norm": 175.12493896484375,
|
| 1163 |
+
"learning_rate": 1.5234105044920526e-05,
|
| 1164 |
+
"loss": 20.55838623046875,
|
| 1165 |
+
"step": 8250
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.704875744418849,
|
| 1169 |
+
"grad_norm": 146.61241149902344,
|
| 1170 |
+
"learning_rate": 1.5018140981340706e-05,
|
| 1171 |
+
"loss": 20.462938232421877,
|
| 1172 |
+
"step": 8300
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.709121983843059,
|
| 1176 |
+
"grad_norm": 132.65420532226562,
|
| 1177 |
+
"learning_rate": 1.4802176917760884e-05,
|
| 1178 |
+
"loss": 20.41062255859375,
|
| 1179 |
+
"step": 8350
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.7133682232672689,
|
| 1183 |
+
"grad_norm": 145.90028381347656,
|
| 1184 |
+
"learning_rate": 1.4586212854181064e-05,
|
| 1185 |
+
"loss": 20.371319580078126,
|
| 1186 |
+
"step": 8400
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.7176144626914789,
|
| 1190 |
+
"grad_norm": 147.31033325195312,
|
| 1191 |
+
"learning_rate": 1.4370248790601245e-05,
|
| 1192 |
+
"loss": 20.66120361328125,
|
| 1193 |
+
"step": 8450
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.7218607021156888,
|
| 1197 |
+
"grad_norm": 136.2473602294922,
|
| 1198 |
+
"learning_rate": 1.4154284727021425e-05,
|
| 1199 |
+
"loss": 20.56037109375,
|
| 1200 |
+
"step": 8500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.7261069415398987,
|
| 1204 |
+
"grad_norm": 152.7163848876953,
|
| 1205 |
+
"learning_rate": 1.3938320663441603e-05,
|
| 1206 |
+
"loss": 20.41483154296875,
|
| 1207 |
+
"step": 8550
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.7303531809641086,
|
| 1211 |
+
"grad_norm": 144.92152404785156,
|
| 1212 |
+
"learning_rate": 1.3722356599861783e-05,
|
| 1213 |
+
"loss": 21.3026416015625,
|
| 1214 |
+
"step": 8600
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.7345994203883186,
|
| 1218 |
+
"grad_norm": 146.7015380859375,
|
| 1219 |
+
"learning_rate": 1.3506392536281965e-05,
|
| 1220 |
+
"loss": 20.11982177734375,
|
| 1221 |
+
"step": 8650
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.7388456598125285,
|
| 1225 |
+
"grad_norm": 122.79088592529297,
|
| 1226 |
+
"learning_rate": 1.3290428472702143e-05,
|
| 1227 |
+
"loss": 20.265406494140624,
|
| 1228 |
+
"step": 8700
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.7430918992367385,
|
| 1232 |
+
"grad_norm": 124.77399444580078,
|
| 1233 |
+
"learning_rate": 1.3074464409122322e-05,
|
| 1234 |
+
"loss": 20.56513427734375,
|
| 1235 |
+
"step": 8750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.7473381386609484,
|
| 1239 |
+
"grad_norm": 131.1453094482422,
|
| 1240 |
+
"learning_rate": 1.28585003455425e-05,
|
| 1241 |
+
"loss": 19.999232177734374,
|
| 1242 |
+
"step": 8800
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.7515843780851583,
|
| 1246 |
+
"grad_norm": 143.78489685058594,
|
| 1247 |
+
"learning_rate": 1.2642536281962684e-05,
|
| 1248 |
+
"loss": 20.73391845703125,
|
| 1249 |
+
"step": 8850
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.7558306175093683,
|
| 1253 |
+
"grad_norm": 144.25245666503906,
|
| 1254 |
+
"learning_rate": 1.2426572218382862e-05,
|
| 1255 |
+
"loss": 20.422523193359375,
|
| 1256 |
+
"step": 8900
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.7600768569335782,
|
| 1260 |
+
"grad_norm": 173.52023315429688,
|
| 1261 |
+
"learning_rate": 1.2210608154803042e-05,
|
| 1262 |
+
"loss": 19.5886962890625,
|
| 1263 |
+
"step": 8950
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.7643230963577882,
|
| 1267 |
+
"grad_norm": 138.1155242919922,
|
| 1268 |
+
"learning_rate": 1.1994644091223221e-05,
|
| 1269 |
+
"loss": 19.826878662109376,
|
| 1270 |
+
"step": 9000
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.768569335781998,
|
| 1274 |
+
"grad_norm": 141.95608520507812,
|
| 1275 |
+
"learning_rate": 1.17786800276434e-05,
|
| 1276 |
+
"loss": 20.6840185546875,
|
| 1277 |
+
"step": 9050
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.772815575206208,
|
| 1281 |
+
"grad_norm": 130.37266540527344,
|
| 1282 |
+
"learning_rate": 1.1562715964063581e-05,
|
| 1283 |
+
"loss": 20.7981884765625,
|
| 1284 |
+
"step": 9100
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.7770618146304179,
|
| 1288 |
+
"grad_norm": 129.22499084472656,
|
| 1289 |
+
"learning_rate": 1.1346751900483759e-05,
|
| 1290 |
+
"loss": 20.298045654296875,
|
| 1291 |
+
"step": 9150
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.7813080540546279,
|
| 1295 |
+
"grad_norm": 196.2136993408203,
|
| 1296 |
+
"learning_rate": 1.113078783690394e-05,
|
| 1297 |
+
"loss": 20.010155029296875,
|
| 1298 |
+
"step": 9200
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.7855542934788378,
|
| 1302 |
+
"grad_norm": 143.95925903320312,
|
| 1303 |
+
"learning_rate": 1.0914823773324119e-05,
|
| 1304 |
+
"loss": 20.69479248046875,
|
| 1305 |
+
"step": 9250
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.7898005329030477,
|
| 1309 |
+
"grad_norm": 140.12205505371094,
|
| 1310 |
+
"learning_rate": 1.06988597097443e-05,
|
| 1311 |
+
"loss": 20.101761474609376,
|
| 1312 |
+
"step": 9300
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.7940467723272576,
|
| 1316 |
+
"grad_norm": 148.62911987304688,
|
| 1317 |
+
"learning_rate": 1.0482895646164478e-05,
|
| 1318 |
+
"loss": 20.48222900390625,
|
| 1319 |
+
"step": 9350
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.7982930117514676,
|
| 1323 |
+
"grad_norm": 134.18821716308594,
|
| 1324 |
+
"learning_rate": 1.0266931582584658e-05,
|
| 1325 |
+
"loss": 21.5114208984375,
|
| 1326 |
+
"step": 9400
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.8025392511756776,
|
| 1330 |
+
"grad_norm": 112.08241271972656,
|
| 1331 |
+
"learning_rate": 1.0050967519004838e-05,
|
| 1332 |
+
"loss": 19.884307861328125,
|
| 1333 |
+
"step": 9450
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.8067854905998875,
|
| 1337 |
+
"grad_norm": 130.55465698242188,
|
| 1338 |
+
"learning_rate": 9.835003455425018e-06,
|
| 1339 |
+
"loss": 19.306004638671876,
|
| 1340 |
+
"step": 9500
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.8110317300240975,
|
| 1344 |
+
"grad_norm": 175.3483428955078,
|
| 1345 |
+
"learning_rate": 9.619039391845197e-06,
|
| 1346 |
+
"loss": 20.79504638671875,
|
| 1347 |
+
"step": 9550
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.8152779694483073,
|
| 1351 |
+
"grad_norm": 123.85011291503906,
|
| 1352 |
+
"learning_rate": 9.403075328265377e-06,
|
| 1353 |
+
"loss": 19.556348876953123,
|
| 1354 |
+
"step": 9600
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.8195242088725173,
|
| 1358 |
+
"grad_norm": 140.8203125,
|
| 1359 |
+
"learning_rate": 9.187111264685557e-06,
|
| 1360 |
+
"loss": 20.43146484375,
|
| 1361 |
+
"step": 9650
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.8237704482967272,
|
| 1365 |
+
"grad_norm": 136.66036987304688,
|
| 1366 |
+
"learning_rate": 8.971147201105737e-06,
|
| 1367 |
+
"loss": 20.33073486328125,
|
| 1368 |
+
"step": 9700
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.8280166877209372,
|
| 1372 |
+
"grad_norm": 146.3016815185547,
|
| 1373 |
+
"learning_rate": 8.755183137525917e-06,
|
| 1374 |
+
"loss": 19.646549072265625,
|
| 1375 |
+
"step": 9750
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.832262927145147,
|
| 1379 |
+
"grad_norm": 148.30975341796875,
|
| 1380 |
+
"learning_rate": 8.539219073946096e-06,
|
| 1381 |
+
"loss": 19.05823486328125,
|
| 1382 |
+
"step": 9800
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.836509166569357,
|
| 1386 |
+
"grad_norm": 140.53770446777344,
|
| 1387 |
+
"learning_rate": 8.323255010366274e-06,
|
| 1388 |
+
"loss": 19.65471923828125,
|
| 1389 |
+
"step": 9850
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.8407554059935669,
|
| 1393 |
+
"grad_norm": 137.27951049804688,
|
| 1394 |
+
"learning_rate": 8.107290946786456e-06,
|
| 1395 |
+
"loss": 20.391009521484374,
|
| 1396 |
+
"step": 9900
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.8450016454177769,
|
| 1400 |
+
"grad_norm": 151.11325073242188,
|
| 1401 |
+
"learning_rate": 7.891326883206634e-06,
|
| 1402 |
+
"loss": 19.69550048828125,
|
| 1403 |
+
"step": 9950
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.8492478848419868,
|
| 1407 |
+
"grad_norm": 137.7413787841797,
|
| 1408 |
+
"learning_rate": 7.675362819626816e-06,
|
| 1409 |
+
"loss": 19.3516943359375,
|
| 1410 |
+
"step": 10000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.8534941242661968,
|
| 1414 |
+
"grad_norm": 112.80725860595703,
|
| 1415 |
+
"learning_rate": 7.459398756046994e-06,
|
| 1416 |
+
"loss": 19.201815185546874,
|
| 1417 |
+
"step": 10050
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.8577403636904067,
|
| 1421 |
+
"grad_norm": 150.98226928710938,
|
| 1422 |
+
"learning_rate": 7.243434692467174e-06,
|
| 1423 |
+
"loss": 19.6190673828125,
|
| 1424 |
+
"step": 10100
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.8619866031146166,
|
| 1428 |
+
"grad_norm": 151.21493530273438,
|
| 1429 |
+
"learning_rate": 7.027470628887353e-06,
|
| 1430 |
+
"loss": 20.10290771484375,
|
| 1431 |
+
"step": 10150
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.8662328425388266,
|
| 1435 |
+
"grad_norm": 166.5272674560547,
|
| 1436 |
+
"learning_rate": 6.811506565307534e-06,
|
| 1437 |
+
"loss": 20.10521728515625,
|
| 1438 |
+
"step": 10200
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.8704790819630365,
|
| 1442 |
+
"grad_norm": 126.49415588378906,
|
| 1443 |
+
"learning_rate": 6.595542501727713e-06,
|
| 1444 |
+
"loss": 19.93487548828125,
|
| 1445 |
+
"step": 10250
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.8747253213872465,
|
| 1449 |
+
"grad_norm": 130.64857482910156,
|
| 1450 |
+
"learning_rate": 6.379578438147893e-06,
|
| 1451 |
+
"loss": 19.3108447265625,
|
| 1452 |
+
"step": 10300
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.8789715608114563,
|
| 1456 |
+
"grad_norm": 155.81996154785156,
|
| 1457 |
+
"learning_rate": 6.1636143745680715e-06,
|
| 1458 |
+
"loss": 19.91413330078125,
|
| 1459 |
+
"step": 10350
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.8832178002356663,
|
| 1463 |
+
"grad_norm": 158.0801544189453,
|
| 1464 |
+
"learning_rate": 5.947650310988251e-06,
|
| 1465 |
+
"loss": 19.63139892578125,
|
| 1466 |
+
"step": 10400
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.8874640396598762,
|
| 1470 |
+
"grad_norm": 138.51763916015625,
|
| 1471 |
+
"learning_rate": 5.731686247408431e-06,
|
| 1472 |
+
"loss": 19.44971923828125,
|
| 1473 |
+
"step": 10450
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.8917102790840862,
|
| 1477 |
+
"grad_norm": 200.97584533691406,
|
| 1478 |
+
"learning_rate": 5.515722183828611e-06,
|
| 1479 |
+
"loss": 19.995697021484375,
|
| 1480 |
+
"step": 10500
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.895956518508296,
|
| 1484 |
+
"grad_norm": 125.70748901367188,
|
| 1485 |
+
"learning_rate": 5.299758120248791e-06,
|
| 1486 |
+
"loss": 19.891861572265626,
|
| 1487 |
+
"step": 10550
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.900202757932506,
|
| 1491 |
+
"grad_norm": 175.50962829589844,
|
| 1492 |
+
"learning_rate": 5.0837940566689705e-06,
|
| 1493 |
+
"loss": 19.681768798828124,
|
| 1494 |
+
"step": 10600
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.9044489973567159,
|
| 1498 |
+
"grad_norm": 129.02464294433594,
|
| 1499 |
+
"learning_rate": 4.86782999308915e-06,
|
| 1500 |
+
"loss": 19.740001220703125,
|
| 1501 |
+
"step": 10650
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.9086952367809259,
|
| 1505 |
+
"grad_norm": 170.73489379882812,
|
| 1506 |
+
"learning_rate": 4.651865929509329e-06,
|
| 1507 |
+
"loss": 20.2353564453125,
|
| 1508 |
+
"step": 10700
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.9129414762051358,
|
| 1512 |
+
"grad_norm": 141.15484619140625,
|
| 1513 |
+
"learning_rate": 4.435901865929509e-06,
|
| 1514 |
+
"loss": 19.27646240234375,
|
| 1515 |
+
"step": 10750
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.9171877156293458,
|
| 1519 |
+
"grad_norm": 139.16778564453125,
|
| 1520 |
+
"learning_rate": 4.219937802349689e-06,
|
| 1521 |
+
"loss": 20.000045166015624,
|
| 1522 |
+
"step": 10800
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.9214339550535557,
|
| 1526 |
+
"grad_norm": 160.3579559326172,
|
| 1527 |
+
"learning_rate": 4.003973738769869e-06,
|
| 1528 |
+
"loss": 18.942919921875,
|
| 1529 |
+
"step": 10850
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.9256801944777656,
|
| 1533 |
+
"grad_norm": 133.4417724609375,
|
| 1534 |
+
"learning_rate": 3.7880096751900484e-06,
|
| 1535 |
+
"loss": 19.30787841796875,
|
| 1536 |
+
"step": 10900
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.9299264339019756,
|
| 1540 |
+
"grad_norm": 147.41517639160156,
|
| 1541 |
+
"learning_rate": 3.572045611610228e-06,
|
| 1542 |
+
"loss": 19.907716064453126,
|
| 1543 |
+
"step": 10950
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.9341726733261855,
|
| 1547 |
+
"grad_norm": 137.2278289794922,
|
| 1548 |
+
"learning_rate": 3.3560815480304076e-06,
|
| 1549 |
+
"loss": 19.394835205078124,
|
| 1550 |
+
"step": 11000
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.9384189127503955,
|
| 1554 |
+
"grad_norm": 148.608642578125,
|
| 1555 |
+
"learning_rate": 3.1401174844505873e-06,
|
| 1556 |
+
"loss": 19.870865478515626,
|
| 1557 |
+
"step": 11050
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.9426651521746053,
|
| 1561 |
+
"grad_norm": 148.3636016845703,
|
| 1562 |
+
"learning_rate": 2.9241534208707676e-06,
|
| 1563 |
+
"loss": 20.17911376953125,
|
| 1564 |
+
"step": 11100
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.9469113915988153,
|
| 1568 |
+
"grad_norm": 136.2633056640625,
|
| 1569 |
+
"learning_rate": 2.708189357290947e-06,
|
| 1570 |
+
"loss": 19.6819775390625,
|
| 1571 |
+
"step": 11150
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.9511576310230252,
|
| 1575 |
+
"grad_norm": 132.42396545410156,
|
| 1576 |
+
"learning_rate": 2.4922252937111267e-06,
|
| 1577 |
+
"loss": 19.143297119140627,
|
| 1578 |
+
"step": 11200
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.9554038704472352,
|
| 1582 |
+
"grad_norm": 137.15524291992188,
|
| 1583 |
+
"learning_rate": 2.2762612301313065e-06,
|
| 1584 |
+
"loss": 19.832156982421875,
|
| 1585 |
+
"step": 11250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.9596501098714451,
|
| 1589 |
+
"grad_norm": 145.9365234375,
|
| 1590 |
+
"learning_rate": 2.060297166551486e-06,
|
| 1591 |
+
"loss": 19.251436767578124,
|
| 1592 |
+
"step": 11300
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.963896349295655,
|
| 1596 |
+
"grad_norm": 142.9067840576172,
|
| 1597 |
+
"learning_rate": 1.8443331029716657e-06,
|
| 1598 |
+
"loss": 19.578409423828123,
|
| 1599 |
+
"step": 11350
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.9681425887198649,
|
| 1603 |
+
"grad_norm": 160.01657104492188,
|
| 1604 |
+
"learning_rate": 1.6283690393918455e-06,
|
| 1605 |
+
"loss": 19.1957666015625,
|
| 1606 |
+
"step": 11400
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.9723888281440749,
|
| 1610 |
+
"grad_norm": 138.53872680664062,
|
| 1611 |
+
"learning_rate": 1.4124049758120248e-06,
|
| 1612 |
+
"loss": 18.92045654296875,
|
| 1613 |
+
"step": 11450
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.9766350675682849,
|
| 1617 |
+
"grad_norm": 150.38726806640625,
|
| 1618 |
+
"learning_rate": 1.1964409122322046e-06,
|
| 1619 |
+
"loss": 19.44462890625,
|
| 1620 |
+
"step": 11500
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.9808813069924948,
|
| 1624 |
+
"grad_norm": 174.97015380859375,
|
| 1625 |
+
"learning_rate": 9.804768486523842e-07,
|
| 1626 |
+
"loss": 19.37765869140625,
|
| 1627 |
+
"step": 11550
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.9851275464167047,
|
| 1631 |
+
"grad_norm": 171.67678833007812,
|
| 1632 |
+
"learning_rate": 7.64512785072564e-07,
|
| 1633 |
+
"loss": 19.446109619140625,
|
| 1634 |
+
"step": 11600
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.9893737858409146,
|
| 1638 |
+
"grad_norm": 163.58985900878906,
|
| 1639 |
+
"learning_rate": 5.485487214927436e-07,
|
| 1640 |
+
"loss": 18.67140625,
|
| 1641 |
+
"step": 11650
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.9936200252651246,
|
| 1645 |
+
"grad_norm": 128.6081085205078,
|
| 1646 |
+
"learning_rate": 3.325846579129233e-07,
|
| 1647 |
+
"loss": 19.808094482421875,
|
| 1648 |
+
"step": 11700
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.9978662646893345,
|
| 1652 |
+
"grad_norm": 132.29139709472656,
|
| 1653 |
+
"learning_rate": 1.1662059433310296e-07,
|
| 1654 |
+
"loss": 19.043531494140623,
|
| 1655 |
+
"step": 11750
|
| 1656 |
+
}
|
| 1657 |
+
],
|
| 1658 |
+
"logging_steps": 50,
|
| 1659 |
+
"max_steps": 11776,
|
| 1660 |
+
"num_input_tokens_seen": 0,
|
| 1661 |
+
"num_train_epochs": 1,
|
| 1662 |
+
"save_steps": 500,
|
| 1663 |
+
"stateful_callbacks": {
|
| 1664 |
+
"TrainerControl": {
|
| 1665 |
+
"args": {
|
| 1666 |
+
"should_epoch_stop": false,
|
| 1667 |
+
"should_evaluate": false,
|
| 1668 |
+
"should_log": false,
|
| 1669 |
+
"should_save": true,
|
| 1670 |
+
"should_training_stop": true
|
| 1671 |
+
},
|
| 1672 |
+
"attributes": {}
|
| 1673 |
+
}
|
| 1674 |
+
},
|
| 1675 |
+
"total_flos": 3.43911674761974e+16,
|
| 1676 |
+
"train_batch_size": 2,
|
| 1677 |
+
"trial_name": null,
|
| 1678 |
+
"trial_params": null
|
| 1679 |
+
}
|
checkpoint-11776/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:38a7040df54471eb85e7c60e05f231ca54e9878768f7e376b60a2a2f85c48eae
|
| 3 |
+
size 5137
|
config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_cross_attention": false,
|
| 3 |
+
"architectures": [
|
| 4 |
+
"XLMRobertaForMaskedLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"dtype": "float32",
|
| 10 |
+
"eos_token_id": 2,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.1,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"is_decoder": false,
|
| 17 |
+
"layer_norm_eps": 1e-05,
|
| 18 |
+
"max_position_embeddings": 514,
|
| 19 |
+
"model_type": "xlm-roberta",
|
| 20 |
+
"num_attention_heads": 12,
|
| 21 |
+
"num_hidden_layers": 12,
|
| 22 |
+
"output_past": true,
|
| 23 |
+
"pad_token_id": 1,
|
| 24 |
+
"position_embedding_type": "absolute",
|
| 25 |
+
"tie_word_embeddings": true,
|
| 26 |
+
"transformers_version": "5.10.2",
|
| 27 |
+
"type_vocab_size": 1,
|
| 28 |
+
"use_cache": false,
|
| 29 |
+
"vocab_size": 250002
|
| 30 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d1d2be64d5ac753fa2bbeda1f7b419a3efb6365ca51e5ec6ab3c653872bfb8a1
|
| 3 |
+
size 1113205088
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:acbd420e2269cdc1ef45332d3d5c418be4aef6b8cb5a0b7ccae0893485307153
|
| 3 |
+
size 17098086
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": true,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<s>",
|
| 5 |
+
"cls_token": "<s>",
|
| 6 |
+
"eos_token": "</s>",
|
| 7 |
+
"is_local": false,
|
| 8 |
+
"local_files_only": false,
|
| 9 |
+
"mask_token": "<mask>",
|
| 10 |
+
"model_max_length": 512,
|
| 11 |
+
"pad_token": "<pad>",
|
| 12 |
+
"sep_token": "</s>",
|
| 13 |
+
"tokenizer_class": "XLMRobertaTokenizer",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:38a7040df54471eb85e7c60e05f231ca54e9878768f7e376b60a2a2f85c48eae
|
| 3 |
+
size 5137
|