Training in progress, step 5000, checkpoint
Browse files- last-checkpoint/config.json +47 -0
- last-checkpoint/model.safetensors +3 -0
- last-checkpoint/optimizer.pt +3 -0
- last-checkpoint/rng_state_0.pth +3 -0
- last-checkpoint/rng_state_1.pth +3 -0
- last-checkpoint/rng_state_2.pth +3 -0
- last-checkpoint/rng_state_3.pth +3 -0
- last-checkpoint/rng_state_4.pth +3 -0
- last-checkpoint/rng_state_5.pth +3 -0
- last-checkpoint/rng_state_6.pth +3 -0
- last-checkpoint/rng_state_7.pth +3 -0
- last-checkpoint/scheduler.pt +3 -0
- last-checkpoint/special_tokens_map.json +37 -0
- last-checkpoint/tokenizer.json +0 -0
- last-checkpoint/tokenizer_config.json +945 -0
- last-checkpoint/trainer_state.json +3541 -0
- last-checkpoint/training_args.bin +3 -0
last-checkpoint/config.json
ADDED
|
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "CocoRoF/KoModernBERT",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"ModernBertForMaskedLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 50281,
|
| 9 |
+
"classifier_activation": "gelu",
|
| 10 |
+
"classifier_bias": false,
|
| 11 |
+
"classifier_dropout": 0.0,
|
| 12 |
+
"classifier_pooling": "mean",
|
| 13 |
+
"cls_token_id": 50281,
|
| 14 |
+
"decoder_bias": true,
|
| 15 |
+
"deterministic_flash_attn": false,
|
| 16 |
+
"embedding_dropout": 0.0,
|
| 17 |
+
"eos_token_id": 50282,
|
| 18 |
+
"global_attn_every_n_layers": 3,
|
| 19 |
+
"global_rope_theta": 160000.0,
|
| 20 |
+
"gradient_checkpointing": false,
|
| 21 |
+
"hidden_activation": "gelu",
|
| 22 |
+
"hidden_size": 768,
|
| 23 |
+
"initializer_cutoff_factor": 2.0,
|
| 24 |
+
"initializer_range": 0.02,
|
| 25 |
+
"intermediate_size": 1152,
|
| 26 |
+
"layer_norm_eps": 1e-05,
|
| 27 |
+
"local_attention": 128,
|
| 28 |
+
"local_rope_theta": 10000.0,
|
| 29 |
+
"max_position_embeddings": 8192,
|
| 30 |
+
"mlp_bias": false,
|
| 31 |
+
"mlp_dropout": 0.0,
|
| 32 |
+
"model_type": "modernbert",
|
| 33 |
+
"norm_bias": false,
|
| 34 |
+
"norm_eps": 1e-05,
|
| 35 |
+
"num_attention_heads": 12,
|
| 36 |
+
"num_hidden_layers": 22,
|
| 37 |
+
"pad_token_id": 50283,
|
| 38 |
+
"position_embedding_type": "absolute",
|
| 39 |
+
"reference_compile": false,
|
| 40 |
+
"repad_logits_with_grad": false,
|
| 41 |
+
"sep_token_id": 50282,
|
| 42 |
+
"sparse_pred_ignore_index": -100,
|
| 43 |
+
"sparse_prediction": false,
|
| 44 |
+
"torch_dtype": "float16",
|
| 45 |
+
"transformers_version": "4.48.1",
|
| 46 |
+
"vocab_size": 55111
|
| 47 |
+
}
|
last-checkpoint/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1c0b904a08084343a53343d611e8f5848ebdac02967f674f70a366d1bde32166
|
| 3 |
+
size 306619286
|
last-checkpoint/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:078d9aa67683190d26f04bb5c7022f94e2cd35b60503efb451eb3289b84cddd0
|
| 3 |
+
size 919972410
|
last-checkpoint/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:78d3f197f6c6558fa8056324f1563ab9e957255f5a1a959362aa4eed7a9545db
|
| 3 |
+
size 15984
|
last-checkpoint/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1c1a9c65c2869356282cad6b4a0f7dff7f4dd68ab3d9d216c72b7d6cb524f860
|
| 3 |
+
size 15984
|
last-checkpoint/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:896febe768e17bae5022a95960c041f6425783774ec8859d99d3b149063b1bf9
|
| 3 |
+
size 15984
|
last-checkpoint/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eac482d57e966585467c8ef44dae2869bf7e5d92886f69c11ed7bccc34c07efe
|
| 3 |
+
size 15984
|
last-checkpoint/rng_state_4.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e1f27d227a20dc320ac283e0938fb2f6e5b475829a583f8c44d1a16a8c828307
|
| 3 |
+
size 15984
|
last-checkpoint/rng_state_5.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d05a7106aaeaec4b81704e3f4a998b5123cf9342a6733bd9fd2d578e99108c3b
|
| 3 |
+
size 15984
|
last-checkpoint/rng_state_6.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b94120d8d88502ec8d8b623ec7550315caca003b44fcffbb5767ab0de91baefe
|
| 3 |
+
size 15984
|
last-checkpoint/rng_state_7.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:332e4d901be380f740b5d8578f7b80ef1865c7fba83bc288c8a35852205cc668
|
| 3 |
+
size 15984
|
last-checkpoint/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3efbe58954d5170ca957430a6bfb4689bc076d59302048b8edf7e3e7b10c40d9
|
| 3 |
+
size 1000
|
last-checkpoint/special_tokens_map.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": {
|
| 3 |
+
"content": "[CLS]",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"mask_token": {
|
| 10 |
+
"content": "[MASK]",
|
| 11 |
+
"lstrip": true,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "[PAD]",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"sep_token": {
|
| 24 |
+
"content": "[SEP]",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"unk_token": {
|
| 31 |
+
"content": "[UNK]",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
}
|
| 37 |
+
}
|
last-checkpoint/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
last-checkpoint/tokenizer_config.json
ADDED
|
@@ -0,0 +1,945 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"50280": {
|
| 4 |
+
"content": "[UNK]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"50281": {
|
| 12 |
+
"content": "[CLS]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"50282": {
|
| 20 |
+
"content": "[SEP]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"50283": {
|
| 28 |
+
"content": "[PAD]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"50284": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": true,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"55000": {
|
| 44 |
+
"content": "|||IP_ADDRESS|||",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": true,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": false
|
| 50 |
+
},
|
| 51 |
+
"55001": {
|
| 52 |
+
"content": "<|padding|>",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
},
|
| 59 |
+
"55002": {
|
| 60 |
+
"content": " ",
|
| 61 |
+
"lstrip": false,
|
| 62 |
+
"normalized": true,
|
| 63 |
+
"rstrip": false,
|
| 64 |
+
"single_word": false,
|
| 65 |
+
"special": false
|
| 66 |
+
},
|
| 67 |
+
"55003": {
|
| 68 |
+
"content": " ",
|
| 69 |
+
"lstrip": false,
|
| 70 |
+
"normalized": true,
|
| 71 |
+
"rstrip": false,
|
| 72 |
+
"single_word": false,
|
| 73 |
+
"special": false
|
| 74 |
+
},
|
| 75 |
+
"55004": {
|
| 76 |
+
"content": " ",
|
| 77 |
+
"lstrip": false,
|
| 78 |
+
"normalized": true,
|
| 79 |
+
"rstrip": false,
|
| 80 |
+
"single_word": false,
|
| 81 |
+
"special": false
|
| 82 |
+
},
|
| 83 |
+
"55005": {
|
| 84 |
+
"content": " ",
|
| 85 |
+
"lstrip": false,
|
| 86 |
+
"normalized": true,
|
| 87 |
+
"rstrip": false,
|
| 88 |
+
"single_word": false,
|
| 89 |
+
"special": false
|
| 90 |
+
},
|
| 91 |
+
"55006": {
|
| 92 |
+
"content": " ",
|
| 93 |
+
"lstrip": false,
|
| 94 |
+
"normalized": true,
|
| 95 |
+
"rstrip": false,
|
| 96 |
+
"single_word": false,
|
| 97 |
+
"special": false
|
| 98 |
+
},
|
| 99 |
+
"55007": {
|
| 100 |
+
"content": " ",
|
| 101 |
+
"lstrip": false,
|
| 102 |
+
"normalized": true,
|
| 103 |
+
"rstrip": false,
|
| 104 |
+
"single_word": false,
|
| 105 |
+
"special": false
|
| 106 |
+
},
|
| 107 |
+
"55008": {
|
| 108 |
+
"content": " ",
|
| 109 |
+
"lstrip": false,
|
| 110 |
+
"normalized": true,
|
| 111 |
+
"rstrip": false,
|
| 112 |
+
"single_word": false,
|
| 113 |
+
"special": false
|
| 114 |
+
},
|
| 115 |
+
"55009": {
|
| 116 |
+
"content": " ",
|
| 117 |
+
"lstrip": false,
|
| 118 |
+
"normalized": true,
|
| 119 |
+
"rstrip": false,
|
| 120 |
+
"single_word": false,
|
| 121 |
+
"special": false
|
| 122 |
+
},
|
| 123 |
+
"55010": {
|
| 124 |
+
"content": " ",
|
| 125 |
+
"lstrip": false,
|
| 126 |
+
"normalized": true,
|
| 127 |
+
"rstrip": false,
|
| 128 |
+
"single_word": false,
|
| 129 |
+
"special": false
|
| 130 |
+
},
|
| 131 |
+
"55011": {
|
| 132 |
+
"content": " ",
|
| 133 |
+
"lstrip": false,
|
| 134 |
+
"normalized": true,
|
| 135 |
+
"rstrip": false,
|
| 136 |
+
"single_word": false,
|
| 137 |
+
"special": false
|
| 138 |
+
},
|
| 139 |
+
"55012": {
|
| 140 |
+
"content": " ",
|
| 141 |
+
"lstrip": false,
|
| 142 |
+
"normalized": true,
|
| 143 |
+
"rstrip": false,
|
| 144 |
+
"single_word": false,
|
| 145 |
+
"special": false
|
| 146 |
+
},
|
| 147 |
+
"55013": {
|
| 148 |
+
"content": " ",
|
| 149 |
+
"lstrip": false,
|
| 150 |
+
"normalized": true,
|
| 151 |
+
"rstrip": false,
|
| 152 |
+
"single_word": false,
|
| 153 |
+
"special": false
|
| 154 |
+
},
|
| 155 |
+
"55014": {
|
| 156 |
+
"content": " ",
|
| 157 |
+
"lstrip": false,
|
| 158 |
+
"normalized": true,
|
| 159 |
+
"rstrip": false,
|
| 160 |
+
"single_word": false,
|
| 161 |
+
"special": false
|
| 162 |
+
},
|
| 163 |
+
"55015": {
|
| 164 |
+
"content": " ",
|
| 165 |
+
"lstrip": false,
|
| 166 |
+
"normalized": true,
|
| 167 |
+
"rstrip": false,
|
| 168 |
+
"single_word": false,
|
| 169 |
+
"special": false
|
| 170 |
+
},
|
| 171 |
+
"55016": {
|
| 172 |
+
"content": " ",
|
| 173 |
+
"lstrip": false,
|
| 174 |
+
"normalized": true,
|
| 175 |
+
"rstrip": false,
|
| 176 |
+
"single_word": false,
|
| 177 |
+
"special": false
|
| 178 |
+
},
|
| 179 |
+
"55017": {
|
| 180 |
+
"content": " ",
|
| 181 |
+
"lstrip": false,
|
| 182 |
+
"normalized": true,
|
| 183 |
+
"rstrip": false,
|
| 184 |
+
"single_word": false,
|
| 185 |
+
"special": false
|
| 186 |
+
},
|
| 187 |
+
"55018": {
|
| 188 |
+
"content": " ",
|
| 189 |
+
"lstrip": false,
|
| 190 |
+
"normalized": true,
|
| 191 |
+
"rstrip": false,
|
| 192 |
+
"single_word": false,
|
| 193 |
+
"special": false
|
| 194 |
+
},
|
| 195 |
+
"55019": {
|
| 196 |
+
"content": " ",
|
| 197 |
+
"lstrip": false,
|
| 198 |
+
"normalized": true,
|
| 199 |
+
"rstrip": false,
|
| 200 |
+
"single_word": false,
|
| 201 |
+
"special": false
|
| 202 |
+
},
|
| 203 |
+
"55020": {
|
| 204 |
+
"content": " ",
|
| 205 |
+
"lstrip": false,
|
| 206 |
+
"normalized": true,
|
| 207 |
+
"rstrip": false,
|
| 208 |
+
"single_word": false,
|
| 209 |
+
"special": false
|
| 210 |
+
},
|
| 211 |
+
"55021": {
|
| 212 |
+
"content": " ",
|
| 213 |
+
"lstrip": false,
|
| 214 |
+
"normalized": true,
|
| 215 |
+
"rstrip": false,
|
| 216 |
+
"single_word": false,
|
| 217 |
+
"special": false
|
| 218 |
+
},
|
| 219 |
+
"55022": {
|
| 220 |
+
"content": " ",
|
| 221 |
+
"lstrip": false,
|
| 222 |
+
"normalized": true,
|
| 223 |
+
"rstrip": false,
|
| 224 |
+
"single_word": false,
|
| 225 |
+
"special": false
|
| 226 |
+
},
|
| 227 |
+
"55023": {
|
| 228 |
+
"content": " ",
|
| 229 |
+
"lstrip": false,
|
| 230 |
+
"normalized": true,
|
| 231 |
+
"rstrip": false,
|
| 232 |
+
"single_word": false,
|
| 233 |
+
"special": false
|
| 234 |
+
},
|
| 235 |
+
"55024": {
|
| 236 |
+
"content": " ",
|
| 237 |
+
"lstrip": false,
|
| 238 |
+
"normalized": true,
|
| 239 |
+
"rstrip": false,
|
| 240 |
+
"single_word": false,
|
| 241 |
+
"special": false
|
| 242 |
+
},
|
| 243 |
+
"55025": {
|
| 244 |
+
"content": "|||EMAIL_ADDRESS|||",
|
| 245 |
+
"lstrip": false,
|
| 246 |
+
"normalized": true,
|
| 247 |
+
"rstrip": false,
|
| 248 |
+
"single_word": false,
|
| 249 |
+
"special": false
|
| 250 |
+
},
|
| 251 |
+
"55026": {
|
| 252 |
+
"content": "|||PHONE_NUMBER|||",
|
| 253 |
+
"lstrip": false,
|
| 254 |
+
"normalized": true,
|
| 255 |
+
"rstrip": false,
|
| 256 |
+
"single_word": false,
|
| 257 |
+
"special": false
|
| 258 |
+
},
|
| 259 |
+
"55027": {
|
| 260 |
+
"content": "<|endoftext|>",
|
| 261 |
+
"lstrip": false,
|
| 262 |
+
"normalized": false,
|
| 263 |
+
"rstrip": false,
|
| 264 |
+
"single_word": false,
|
| 265 |
+
"special": true
|
| 266 |
+
},
|
| 267 |
+
"55028": {
|
| 268 |
+
"content": "[unused0]",
|
| 269 |
+
"lstrip": false,
|
| 270 |
+
"normalized": true,
|
| 271 |
+
"rstrip": false,
|
| 272 |
+
"single_word": false,
|
| 273 |
+
"special": false
|
| 274 |
+
},
|
| 275 |
+
"55029": {
|
| 276 |
+
"content": "[unused1]",
|
| 277 |
+
"lstrip": false,
|
| 278 |
+
"normalized": true,
|
| 279 |
+
"rstrip": false,
|
| 280 |
+
"single_word": false,
|
| 281 |
+
"special": false
|
| 282 |
+
},
|
| 283 |
+
"55030": {
|
| 284 |
+
"content": "[unused2]",
|
| 285 |
+
"lstrip": false,
|
| 286 |
+
"normalized": true,
|
| 287 |
+
"rstrip": false,
|
| 288 |
+
"single_word": false,
|
| 289 |
+
"special": false
|
| 290 |
+
},
|
| 291 |
+
"55031": {
|
| 292 |
+
"content": "[unused3]",
|
| 293 |
+
"lstrip": false,
|
| 294 |
+
"normalized": true,
|
| 295 |
+
"rstrip": false,
|
| 296 |
+
"single_word": false,
|
| 297 |
+
"special": false
|
| 298 |
+
},
|
| 299 |
+
"55032": {
|
| 300 |
+
"content": "[unused4]",
|
| 301 |
+
"lstrip": false,
|
| 302 |
+
"normalized": true,
|
| 303 |
+
"rstrip": false,
|
| 304 |
+
"single_word": false,
|
| 305 |
+
"special": false
|
| 306 |
+
},
|
| 307 |
+
"55033": {
|
| 308 |
+
"content": "[unused5]",
|
| 309 |
+
"lstrip": false,
|
| 310 |
+
"normalized": true,
|
| 311 |
+
"rstrip": false,
|
| 312 |
+
"single_word": false,
|
| 313 |
+
"special": false
|
| 314 |
+
},
|
| 315 |
+
"55034": {
|
| 316 |
+
"content": "[unused6]",
|
| 317 |
+
"lstrip": false,
|
| 318 |
+
"normalized": true,
|
| 319 |
+
"rstrip": false,
|
| 320 |
+
"single_word": false,
|
| 321 |
+
"special": false
|
| 322 |
+
},
|
| 323 |
+
"55035": {
|
| 324 |
+
"content": "[unused7]",
|
| 325 |
+
"lstrip": false,
|
| 326 |
+
"normalized": true,
|
| 327 |
+
"rstrip": false,
|
| 328 |
+
"single_word": false,
|
| 329 |
+
"special": false
|
| 330 |
+
},
|
| 331 |
+
"55036": {
|
| 332 |
+
"content": "[unused8]",
|
| 333 |
+
"lstrip": false,
|
| 334 |
+
"normalized": true,
|
| 335 |
+
"rstrip": false,
|
| 336 |
+
"single_word": false,
|
| 337 |
+
"special": false
|
| 338 |
+
},
|
| 339 |
+
"55037": {
|
| 340 |
+
"content": "[unused9]",
|
| 341 |
+
"lstrip": false,
|
| 342 |
+
"normalized": true,
|
| 343 |
+
"rstrip": false,
|
| 344 |
+
"single_word": false,
|
| 345 |
+
"special": false
|
| 346 |
+
},
|
| 347 |
+
"55038": {
|
| 348 |
+
"content": "[unused10]",
|
| 349 |
+
"lstrip": false,
|
| 350 |
+
"normalized": true,
|
| 351 |
+
"rstrip": false,
|
| 352 |
+
"single_word": false,
|
| 353 |
+
"special": false
|
| 354 |
+
},
|
| 355 |
+
"55039": {
|
| 356 |
+
"content": "[unused11]",
|
| 357 |
+
"lstrip": false,
|
| 358 |
+
"normalized": true,
|
| 359 |
+
"rstrip": false,
|
| 360 |
+
"single_word": false,
|
| 361 |
+
"special": false
|
| 362 |
+
},
|
| 363 |
+
"55040": {
|
| 364 |
+
"content": "[unused12]",
|
| 365 |
+
"lstrip": false,
|
| 366 |
+
"normalized": true,
|
| 367 |
+
"rstrip": false,
|
| 368 |
+
"single_word": false,
|
| 369 |
+
"special": false
|
| 370 |
+
},
|
| 371 |
+
"55041": {
|
| 372 |
+
"content": "[unused13]",
|
| 373 |
+
"lstrip": false,
|
| 374 |
+
"normalized": true,
|
| 375 |
+
"rstrip": false,
|
| 376 |
+
"single_word": false,
|
| 377 |
+
"special": false
|
| 378 |
+
},
|
| 379 |
+
"55042": {
|
| 380 |
+
"content": "[unused14]",
|
| 381 |
+
"lstrip": false,
|
| 382 |
+
"normalized": true,
|
| 383 |
+
"rstrip": false,
|
| 384 |
+
"single_word": false,
|
| 385 |
+
"special": false
|
| 386 |
+
},
|
| 387 |
+
"55043": {
|
| 388 |
+
"content": "[unused15]",
|
| 389 |
+
"lstrip": false,
|
| 390 |
+
"normalized": true,
|
| 391 |
+
"rstrip": false,
|
| 392 |
+
"single_word": false,
|
| 393 |
+
"special": false
|
| 394 |
+
},
|
| 395 |
+
"55044": {
|
| 396 |
+
"content": "[unused16]",
|
| 397 |
+
"lstrip": false,
|
| 398 |
+
"normalized": true,
|
| 399 |
+
"rstrip": false,
|
| 400 |
+
"single_word": false,
|
| 401 |
+
"special": false
|
| 402 |
+
},
|
| 403 |
+
"55045": {
|
| 404 |
+
"content": "[unused17]",
|
| 405 |
+
"lstrip": false,
|
| 406 |
+
"normalized": true,
|
| 407 |
+
"rstrip": false,
|
| 408 |
+
"single_word": false,
|
| 409 |
+
"special": false
|
| 410 |
+
},
|
| 411 |
+
"55046": {
|
| 412 |
+
"content": "[unused18]",
|
| 413 |
+
"lstrip": false,
|
| 414 |
+
"normalized": true,
|
| 415 |
+
"rstrip": false,
|
| 416 |
+
"single_word": false,
|
| 417 |
+
"special": false
|
| 418 |
+
},
|
| 419 |
+
"55047": {
|
| 420 |
+
"content": "[unused19]",
|
| 421 |
+
"lstrip": false,
|
| 422 |
+
"normalized": true,
|
| 423 |
+
"rstrip": false,
|
| 424 |
+
"single_word": false,
|
| 425 |
+
"special": false
|
| 426 |
+
},
|
| 427 |
+
"55048": {
|
| 428 |
+
"content": "[unused20]",
|
| 429 |
+
"lstrip": false,
|
| 430 |
+
"normalized": true,
|
| 431 |
+
"rstrip": false,
|
| 432 |
+
"single_word": false,
|
| 433 |
+
"special": false
|
| 434 |
+
},
|
| 435 |
+
"55049": {
|
| 436 |
+
"content": "[unused21]",
|
| 437 |
+
"lstrip": false,
|
| 438 |
+
"normalized": true,
|
| 439 |
+
"rstrip": false,
|
| 440 |
+
"single_word": false,
|
| 441 |
+
"special": false
|
| 442 |
+
},
|
| 443 |
+
"55050": {
|
| 444 |
+
"content": "[unused22]",
|
| 445 |
+
"lstrip": false,
|
| 446 |
+
"normalized": true,
|
| 447 |
+
"rstrip": false,
|
| 448 |
+
"single_word": false,
|
| 449 |
+
"special": false
|
| 450 |
+
},
|
| 451 |
+
"55051": {
|
| 452 |
+
"content": "[unused23]",
|
| 453 |
+
"lstrip": false,
|
| 454 |
+
"normalized": true,
|
| 455 |
+
"rstrip": false,
|
| 456 |
+
"single_word": false,
|
| 457 |
+
"special": false
|
| 458 |
+
},
|
| 459 |
+
"55052": {
|
| 460 |
+
"content": "[unused24]",
|
| 461 |
+
"lstrip": false,
|
| 462 |
+
"normalized": true,
|
| 463 |
+
"rstrip": false,
|
| 464 |
+
"single_word": false,
|
| 465 |
+
"special": false
|
| 466 |
+
},
|
| 467 |
+
"55053": {
|
| 468 |
+
"content": "[unused25]",
|
| 469 |
+
"lstrip": false,
|
| 470 |
+
"normalized": true,
|
| 471 |
+
"rstrip": false,
|
| 472 |
+
"single_word": false,
|
| 473 |
+
"special": false
|
| 474 |
+
},
|
| 475 |
+
"55054": {
|
| 476 |
+
"content": "[unused26]",
|
| 477 |
+
"lstrip": false,
|
| 478 |
+
"normalized": true,
|
| 479 |
+
"rstrip": false,
|
| 480 |
+
"single_word": false,
|
| 481 |
+
"special": false
|
| 482 |
+
},
|
| 483 |
+
"55055": {
|
| 484 |
+
"content": "[unused27]",
|
| 485 |
+
"lstrip": false,
|
| 486 |
+
"normalized": true,
|
| 487 |
+
"rstrip": false,
|
| 488 |
+
"single_word": false,
|
| 489 |
+
"special": false
|
| 490 |
+
},
|
| 491 |
+
"55056": {
|
| 492 |
+
"content": "[unused28]",
|
| 493 |
+
"lstrip": false,
|
| 494 |
+
"normalized": true,
|
| 495 |
+
"rstrip": false,
|
| 496 |
+
"single_word": false,
|
| 497 |
+
"special": false
|
| 498 |
+
},
|
| 499 |
+
"55057": {
|
| 500 |
+
"content": "[unused29]",
|
| 501 |
+
"lstrip": false,
|
| 502 |
+
"normalized": true,
|
| 503 |
+
"rstrip": false,
|
| 504 |
+
"single_word": false,
|
| 505 |
+
"special": false
|
| 506 |
+
},
|
| 507 |
+
"55058": {
|
| 508 |
+
"content": "[unused30]",
|
| 509 |
+
"lstrip": false,
|
| 510 |
+
"normalized": true,
|
| 511 |
+
"rstrip": false,
|
| 512 |
+
"single_word": false,
|
| 513 |
+
"special": false
|
| 514 |
+
},
|
| 515 |
+
"55059": {
|
| 516 |
+
"content": "[unused31]",
|
| 517 |
+
"lstrip": false,
|
| 518 |
+
"normalized": true,
|
| 519 |
+
"rstrip": false,
|
| 520 |
+
"single_word": false,
|
| 521 |
+
"special": false
|
| 522 |
+
},
|
| 523 |
+
"55060": {
|
| 524 |
+
"content": "[unused32]",
|
| 525 |
+
"lstrip": false,
|
| 526 |
+
"normalized": true,
|
| 527 |
+
"rstrip": false,
|
| 528 |
+
"single_word": false,
|
| 529 |
+
"special": false
|
| 530 |
+
},
|
| 531 |
+
"55061": {
|
| 532 |
+
"content": "[unused33]",
|
| 533 |
+
"lstrip": false,
|
| 534 |
+
"normalized": true,
|
| 535 |
+
"rstrip": false,
|
| 536 |
+
"single_word": false,
|
| 537 |
+
"special": false
|
| 538 |
+
},
|
| 539 |
+
"55062": {
|
| 540 |
+
"content": "[unused34]",
|
| 541 |
+
"lstrip": false,
|
| 542 |
+
"normalized": true,
|
| 543 |
+
"rstrip": false,
|
| 544 |
+
"single_word": false,
|
| 545 |
+
"special": false
|
| 546 |
+
},
|
| 547 |
+
"55063": {
|
| 548 |
+
"content": "[unused35]",
|
| 549 |
+
"lstrip": false,
|
| 550 |
+
"normalized": true,
|
| 551 |
+
"rstrip": false,
|
| 552 |
+
"single_word": false,
|
| 553 |
+
"special": false
|
| 554 |
+
},
|
| 555 |
+
"55064": {
|
| 556 |
+
"content": "[unused36]",
|
| 557 |
+
"lstrip": false,
|
| 558 |
+
"normalized": true,
|
| 559 |
+
"rstrip": false,
|
| 560 |
+
"single_word": false,
|
| 561 |
+
"special": false
|
| 562 |
+
},
|
| 563 |
+
"55065": {
|
| 564 |
+
"content": "[unused37]",
|
| 565 |
+
"lstrip": false,
|
| 566 |
+
"normalized": true,
|
| 567 |
+
"rstrip": false,
|
| 568 |
+
"single_word": false,
|
| 569 |
+
"special": false
|
| 570 |
+
},
|
| 571 |
+
"55066": {
|
| 572 |
+
"content": "[unused38]",
|
| 573 |
+
"lstrip": false,
|
| 574 |
+
"normalized": true,
|
| 575 |
+
"rstrip": false,
|
| 576 |
+
"single_word": false,
|
| 577 |
+
"special": false
|
| 578 |
+
},
|
| 579 |
+
"55067": {
|
| 580 |
+
"content": "[unused39]",
|
| 581 |
+
"lstrip": false,
|
| 582 |
+
"normalized": true,
|
| 583 |
+
"rstrip": false,
|
| 584 |
+
"single_word": false,
|
| 585 |
+
"special": false
|
| 586 |
+
},
|
| 587 |
+
"55068": {
|
| 588 |
+
"content": "[unused40]",
|
| 589 |
+
"lstrip": false,
|
| 590 |
+
"normalized": true,
|
| 591 |
+
"rstrip": false,
|
| 592 |
+
"single_word": false,
|
| 593 |
+
"special": false
|
| 594 |
+
},
|
| 595 |
+
"55069": {
|
| 596 |
+
"content": "[unused41]",
|
| 597 |
+
"lstrip": false,
|
| 598 |
+
"normalized": true,
|
| 599 |
+
"rstrip": false,
|
| 600 |
+
"single_word": false,
|
| 601 |
+
"special": false
|
| 602 |
+
},
|
| 603 |
+
"55070": {
|
| 604 |
+
"content": "[unused42]",
|
| 605 |
+
"lstrip": false,
|
| 606 |
+
"normalized": true,
|
| 607 |
+
"rstrip": false,
|
| 608 |
+
"single_word": false,
|
| 609 |
+
"special": false
|
| 610 |
+
},
|
| 611 |
+
"55071": {
|
| 612 |
+
"content": "[unused43]",
|
| 613 |
+
"lstrip": false,
|
| 614 |
+
"normalized": true,
|
| 615 |
+
"rstrip": false,
|
| 616 |
+
"single_word": false,
|
| 617 |
+
"special": false
|
| 618 |
+
},
|
| 619 |
+
"55072": {
|
| 620 |
+
"content": "[unused44]",
|
| 621 |
+
"lstrip": false,
|
| 622 |
+
"normalized": true,
|
| 623 |
+
"rstrip": false,
|
| 624 |
+
"single_word": false,
|
| 625 |
+
"special": false
|
| 626 |
+
},
|
| 627 |
+
"55073": {
|
| 628 |
+
"content": "[unused45]",
|
| 629 |
+
"lstrip": false,
|
| 630 |
+
"normalized": true,
|
| 631 |
+
"rstrip": false,
|
| 632 |
+
"single_word": false,
|
| 633 |
+
"special": false
|
| 634 |
+
},
|
| 635 |
+
"55074": {
|
| 636 |
+
"content": "[unused46]",
|
| 637 |
+
"lstrip": false,
|
| 638 |
+
"normalized": true,
|
| 639 |
+
"rstrip": false,
|
| 640 |
+
"single_word": false,
|
| 641 |
+
"special": false
|
| 642 |
+
},
|
| 643 |
+
"55075": {
|
| 644 |
+
"content": "[unused47]",
|
| 645 |
+
"lstrip": false,
|
| 646 |
+
"normalized": true,
|
| 647 |
+
"rstrip": false,
|
| 648 |
+
"single_word": false,
|
| 649 |
+
"special": false
|
| 650 |
+
},
|
| 651 |
+
"55076": {
|
| 652 |
+
"content": "[unused48]",
|
| 653 |
+
"lstrip": false,
|
| 654 |
+
"normalized": true,
|
| 655 |
+
"rstrip": false,
|
| 656 |
+
"single_word": false,
|
| 657 |
+
"special": false
|
| 658 |
+
},
|
| 659 |
+
"55077": {
|
| 660 |
+
"content": "[unused49]",
|
| 661 |
+
"lstrip": false,
|
| 662 |
+
"normalized": true,
|
| 663 |
+
"rstrip": false,
|
| 664 |
+
"single_word": false,
|
| 665 |
+
"special": false
|
| 666 |
+
},
|
| 667 |
+
"55078": {
|
| 668 |
+
"content": "[unused50]",
|
| 669 |
+
"lstrip": false,
|
| 670 |
+
"normalized": true,
|
| 671 |
+
"rstrip": false,
|
| 672 |
+
"single_word": false,
|
| 673 |
+
"special": false
|
| 674 |
+
},
|
| 675 |
+
"55079": {
|
| 676 |
+
"content": "[unused51]",
|
| 677 |
+
"lstrip": false,
|
| 678 |
+
"normalized": true,
|
| 679 |
+
"rstrip": false,
|
| 680 |
+
"single_word": false,
|
| 681 |
+
"special": false
|
| 682 |
+
},
|
| 683 |
+
"55080": {
|
| 684 |
+
"content": "[unused52]",
|
| 685 |
+
"lstrip": false,
|
| 686 |
+
"normalized": true,
|
| 687 |
+
"rstrip": false,
|
| 688 |
+
"single_word": false,
|
| 689 |
+
"special": false
|
| 690 |
+
},
|
| 691 |
+
"55081": {
|
| 692 |
+
"content": "[unused53]",
|
| 693 |
+
"lstrip": false,
|
| 694 |
+
"normalized": true,
|
| 695 |
+
"rstrip": false,
|
| 696 |
+
"single_word": false,
|
| 697 |
+
"special": false
|
| 698 |
+
},
|
| 699 |
+
"55082": {
|
| 700 |
+
"content": "[unused54]",
|
| 701 |
+
"lstrip": false,
|
| 702 |
+
"normalized": true,
|
| 703 |
+
"rstrip": false,
|
| 704 |
+
"single_word": false,
|
| 705 |
+
"special": false
|
| 706 |
+
},
|
| 707 |
+
"55083": {
|
| 708 |
+
"content": "[unused55]",
|
| 709 |
+
"lstrip": false,
|
| 710 |
+
"normalized": true,
|
| 711 |
+
"rstrip": false,
|
| 712 |
+
"single_word": false,
|
| 713 |
+
"special": false
|
| 714 |
+
},
|
| 715 |
+
"55084": {
|
| 716 |
+
"content": "[unused56]",
|
| 717 |
+
"lstrip": false,
|
| 718 |
+
"normalized": true,
|
| 719 |
+
"rstrip": false,
|
| 720 |
+
"single_word": false,
|
| 721 |
+
"special": false
|
| 722 |
+
},
|
| 723 |
+
"55085": {
|
| 724 |
+
"content": "[unused57]",
|
| 725 |
+
"lstrip": false,
|
| 726 |
+
"normalized": true,
|
| 727 |
+
"rstrip": false,
|
| 728 |
+
"single_word": false,
|
| 729 |
+
"special": false
|
| 730 |
+
},
|
| 731 |
+
"55086": {
|
| 732 |
+
"content": "[unused58]",
|
| 733 |
+
"lstrip": false,
|
| 734 |
+
"normalized": true,
|
| 735 |
+
"rstrip": false,
|
| 736 |
+
"single_word": false,
|
| 737 |
+
"special": false
|
| 738 |
+
},
|
| 739 |
+
"55087": {
|
| 740 |
+
"content": "[unused59]",
|
| 741 |
+
"lstrip": false,
|
| 742 |
+
"normalized": true,
|
| 743 |
+
"rstrip": false,
|
| 744 |
+
"single_word": false,
|
| 745 |
+
"special": false
|
| 746 |
+
},
|
| 747 |
+
"55088": {
|
| 748 |
+
"content": "[unused60]",
|
| 749 |
+
"lstrip": false,
|
| 750 |
+
"normalized": true,
|
| 751 |
+
"rstrip": false,
|
| 752 |
+
"single_word": false,
|
| 753 |
+
"special": false
|
| 754 |
+
},
|
| 755 |
+
"55089": {
|
| 756 |
+
"content": "[unused61]",
|
| 757 |
+
"lstrip": false,
|
| 758 |
+
"normalized": true,
|
| 759 |
+
"rstrip": false,
|
| 760 |
+
"single_word": false,
|
| 761 |
+
"special": false
|
| 762 |
+
},
|
| 763 |
+
"55090": {
|
| 764 |
+
"content": "[unused62]",
|
| 765 |
+
"lstrip": false,
|
| 766 |
+
"normalized": true,
|
| 767 |
+
"rstrip": false,
|
| 768 |
+
"single_word": false,
|
| 769 |
+
"special": false
|
| 770 |
+
},
|
| 771 |
+
"55091": {
|
| 772 |
+
"content": "[unused63]",
|
| 773 |
+
"lstrip": false,
|
| 774 |
+
"normalized": true,
|
| 775 |
+
"rstrip": false,
|
| 776 |
+
"single_word": false,
|
| 777 |
+
"special": false
|
| 778 |
+
},
|
| 779 |
+
"55092": {
|
| 780 |
+
"content": "[unused64]",
|
| 781 |
+
"lstrip": false,
|
| 782 |
+
"normalized": true,
|
| 783 |
+
"rstrip": false,
|
| 784 |
+
"single_word": false,
|
| 785 |
+
"special": false
|
| 786 |
+
},
|
| 787 |
+
"55093": {
|
| 788 |
+
"content": "[unused65]",
|
| 789 |
+
"lstrip": false,
|
| 790 |
+
"normalized": true,
|
| 791 |
+
"rstrip": false,
|
| 792 |
+
"single_word": false,
|
| 793 |
+
"special": false
|
| 794 |
+
},
|
| 795 |
+
"55094": {
|
| 796 |
+
"content": "[unused66]",
|
| 797 |
+
"lstrip": false,
|
| 798 |
+
"normalized": true,
|
| 799 |
+
"rstrip": false,
|
| 800 |
+
"single_word": false,
|
| 801 |
+
"special": false
|
| 802 |
+
},
|
| 803 |
+
"55095": {
|
| 804 |
+
"content": "[unused67]",
|
| 805 |
+
"lstrip": false,
|
| 806 |
+
"normalized": true,
|
| 807 |
+
"rstrip": false,
|
| 808 |
+
"single_word": false,
|
| 809 |
+
"special": false
|
| 810 |
+
},
|
| 811 |
+
"55096": {
|
| 812 |
+
"content": "[unused68]",
|
| 813 |
+
"lstrip": false,
|
| 814 |
+
"normalized": true,
|
| 815 |
+
"rstrip": false,
|
| 816 |
+
"single_word": false,
|
| 817 |
+
"special": false
|
| 818 |
+
},
|
| 819 |
+
"55097": {
|
| 820 |
+
"content": "[unused69]",
|
| 821 |
+
"lstrip": false,
|
| 822 |
+
"normalized": true,
|
| 823 |
+
"rstrip": false,
|
| 824 |
+
"single_word": false,
|
| 825 |
+
"special": false
|
| 826 |
+
},
|
| 827 |
+
"55098": {
|
| 828 |
+
"content": "[unused70]",
|
| 829 |
+
"lstrip": false,
|
| 830 |
+
"normalized": true,
|
| 831 |
+
"rstrip": false,
|
| 832 |
+
"single_word": false,
|
| 833 |
+
"special": false
|
| 834 |
+
},
|
| 835 |
+
"55099": {
|
| 836 |
+
"content": "[unused71]",
|
| 837 |
+
"lstrip": false,
|
| 838 |
+
"normalized": true,
|
| 839 |
+
"rstrip": false,
|
| 840 |
+
"single_word": false,
|
| 841 |
+
"special": false
|
| 842 |
+
},
|
| 843 |
+
"55100": {
|
| 844 |
+
"content": "[unused72]",
|
| 845 |
+
"lstrip": false,
|
| 846 |
+
"normalized": true,
|
| 847 |
+
"rstrip": false,
|
| 848 |
+
"single_word": false,
|
| 849 |
+
"special": false
|
| 850 |
+
},
|
| 851 |
+
"55101": {
|
| 852 |
+
"content": "[unused73]",
|
| 853 |
+
"lstrip": false,
|
| 854 |
+
"normalized": true,
|
| 855 |
+
"rstrip": false,
|
| 856 |
+
"single_word": false,
|
| 857 |
+
"special": false
|
| 858 |
+
},
|
| 859 |
+
"55102": {
|
| 860 |
+
"content": "[unused74]",
|
| 861 |
+
"lstrip": false,
|
| 862 |
+
"normalized": true,
|
| 863 |
+
"rstrip": false,
|
| 864 |
+
"single_word": false,
|
| 865 |
+
"special": false
|
| 866 |
+
},
|
| 867 |
+
"55103": {
|
| 868 |
+
"content": "[unused75]",
|
| 869 |
+
"lstrip": false,
|
| 870 |
+
"normalized": true,
|
| 871 |
+
"rstrip": false,
|
| 872 |
+
"single_word": false,
|
| 873 |
+
"special": false
|
| 874 |
+
},
|
| 875 |
+
"55104": {
|
| 876 |
+
"content": "[unused76]",
|
| 877 |
+
"lstrip": false,
|
| 878 |
+
"normalized": true,
|
| 879 |
+
"rstrip": false,
|
| 880 |
+
"single_word": false,
|
| 881 |
+
"special": false
|
| 882 |
+
},
|
| 883 |
+
"55105": {
|
| 884 |
+
"content": "[unused77]",
|
| 885 |
+
"lstrip": false,
|
| 886 |
+
"normalized": true,
|
| 887 |
+
"rstrip": false,
|
| 888 |
+
"single_word": false,
|
| 889 |
+
"special": false
|
| 890 |
+
},
|
| 891 |
+
"55106": {
|
| 892 |
+
"content": "[unused78]",
|
| 893 |
+
"lstrip": false,
|
| 894 |
+
"normalized": true,
|
| 895 |
+
"rstrip": false,
|
| 896 |
+
"single_word": false,
|
| 897 |
+
"special": false
|
| 898 |
+
},
|
| 899 |
+
"55107": {
|
| 900 |
+
"content": "[unused79]",
|
| 901 |
+
"lstrip": false,
|
| 902 |
+
"normalized": true,
|
| 903 |
+
"rstrip": false,
|
| 904 |
+
"single_word": false,
|
| 905 |
+
"special": false
|
| 906 |
+
},
|
| 907 |
+
"55108": {
|
| 908 |
+
"content": "[unused80]",
|
| 909 |
+
"lstrip": false,
|
| 910 |
+
"normalized": true,
|
| 911 |
+
"rstrip": false,
|
| 912 |
+
"single_word": false,
|
| 913 |
+
"special": false
|
| 914 |
+
},
|
| 915 |
+
"55109": {
|
| 916 |
+
"content": "[unused81]",
|
| 917 |
+
"lstrip": false,
|
| 918 |
+
"normalized": true,
|
| 919 |
+
"rstrip": false,
|
| 920 |
+
"single_word": false,
|
| 921 |
+
"special": false
|
| 922 |
+
},
|
| 923 |
+
"55110": {
|
| 924 |
+
"content": "[unused82]",
|
| 925 |
+
"lstrip": false,
|
| 926 |
+
"normalized": true,
|
| 927 |
+
"rstrip": false,
|
| 928 |
+
"single_word": false,
|
| 929 |
+
"special": false
|
| 930 |
+
}
|
| 931 |
+
},
|
| 932 |
+
"clean_up_tokenization_spaces": true,
|
| 933 |
+
"cls_token": "[CLS]",
|
| 934 |
+
"extra_special_tokens": {},
|
| 935 |
+
"mask_token": "[MASK]",
|
| 936 |
+
"model_input_names": [
|
| 937 |
+
"input_ids",
|
| 938 |
+
"attention_mask"
|
| 939 |
+
],
|
| 940 |
+
"model_max_length": 8192,
|
| 941 |
+
"pad_token": "[PAD]",
|
| 942 |
+
"sep_token": "[SEP]",
|
| 943 |
+
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 944 |
+
"unk_token": "[UNK]"
|
| 945 |
+
}
|
last-checkpoint/trainer_state.json
ADDED
|
@@ -0,0 +1,3541 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.09278654223991352,
|
| 5 |
+
"eval_steps": 5000,
|
| 6 |
+
"global_step": 5000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.00018557308447982704,
|
| 13 |
+
"grad_norm": 28.265625,
|
| 14 |
+
"learning_rate": 9.999971004213958e-06,
|
| 15 |
+
"loss": 80.2344,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0003711461689596541,
|
| 20 |
+
"grad_norm": 27.53125,
|
| 21 |
+
"learning_rate": 9.999942008427916e-06,
|
| 22 |
+
"loss": 73.1461,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0005567192534394811,
|
| 27 |
+
"grad_norm": 24.703125,
|
| 28 |
+
"learning_rate": 9.999913012641873e-06,
|
| 29 |
+
"loss": 70.992,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0007422923379193082,
|
| 34 |
+
"grad_norm": 24.046875,
|
| 35 |
+
"learning_rate": 9.99988401685583e-06,
|
| 36 |
+
"loss": 69.4784,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0009278654223991352,
|
| 41 |
+
"grad_norm": 23.203125,
|
| 42 |
+
"learning_rate": 9.999855021069788e-06,
|
| 43 |
+
"loss": 67.9841,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.0011134385068789622,
|
| 48 |
+
"grad_norm": 25.015625,
|
| 49 |
+
"learning_rate": 9.999826025283747e-06,
|
| 50 |
+
"loss": 66.8785,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.0012990115913587893,
|
| 55 |
+
"grad_norm": 28.875,
|
| 56 |
+
"learning_rate": 9.999797029497704e-06,
|
| 57 |
+
"loss": 65.5765,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.0014845846758386164,
|
| 62 |
+
"grad_norm": 34.375,
|
| 63 |
+
"learning_rate": 9.999768033711662e-06,
|
| 64 |
+
"loss": 64.6408,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.0016701577603184434,
|
| 69 |
+
"grad_norm": 30.875,
|
| 70 |
+
"learning_rate": 9.99973903792562e-06,
|
| 71 |
+
"loss": 63.1597,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.0018557308447982705,
|
| 76 |
+
"grad_norm": 25.90625,
|
| 77 |
+
"learning_rate": 9.999710042139577e-06,
|
| 78 |
+
"loss": 61.7359,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.0020413039292780976,
|
| 83 |
+
"grad_norm": 27.625,
|
| 84 |
+
"learning_rate": 9.999681046353534e-06,
|
| 85 |
+
"loss": 61.0914,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.0022268770137579244,
|
| 90 |
+
"grad_norm": 26.453125,
|
| 91 |
+
"learning_rate": 9.999652050567491e-06,
|
| 92 |
+
"loss": 59.8145,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.0024124500982377517,
|
| 97 |
+
"grad_norm": 33.40625,
|
| 98 |
+
"learning_rate": 9.99962305478145e-06,
|
| 99 |
+
"loss": 58.7817,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.0025980231827175786,
|
| 104 |
+
"grad_norm": 27.578125,
|
| 105 |
+
"learning_rate": 9.999594058995406e-06,
|
| 106 |
+
"loss": 57.699,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.002783596267197406,
|
| 111 |
+
"grad_norm": 25.59375,
|
| 112 |
+
"learning_rate": 9.999565063209364e-06,
|
| 113 |
+
"loss": 56.9819,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.0029691693516772327,
|
| 118 |
+
"grad_norm": 26.84375,
|
| 119 |
+
"learning_rate": 9.999536067423323e-06,
|
| 120 |
+
"loss": 55.928,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.0031547424361570596,
|
| 125 |
+
"grad_norm": 26.734375,
|
| 126 |
+
"learning_rate": 9.99950707163728e-06,
|
| 127 |
+
"loss": 54.613,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.003340315520636887,
|
| 132 |
+
"grad_norm": 28.75,
|
| 133 |
+
"learning_rate": 9.999478075851238e-06,
|
| 134 |
+
"loss": 53.661,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.0035258886051167137,
|
| 139 |
+
"grad_norm": 30.421875,
|
| 140 |
+
"learning_rate": 9.999449080065195e-06,
|
| 141 |
+
"loss": 51.9486,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.003711461689596541,
|
| 146 |
+
"grad_norm": 29.140625,
|
| 147 |
+
"learning_rate": 9.999420084279152e-06,
|
| 148 |
+
"loss": 51.4946,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.003897034774076368,
|
| 153 |
+
"grad_norm": 37.1875,
|
| 154 |
+
"learning_rate": 9.99939108849311e-06,
|
| 155 |
+
"loss": 50.4218,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.004082607858556195,
|
| 160 |
+
"grad_norm": 32.0,
|
| 161 |
+
"learning_rate": 9.999362092707067e-06,
|
| 162 |
+
"loss": 49.2055,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.004268180943036022,
|
| 167 |
+
"grad_norm": 28.40625,
|
| 168 |
+
"learning_rate": 9.999333096921026e-06,
|
| 169 |
+
"loss": 48.1393,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.004453754027515849,
|
| 174 |
+
"grad_norm": 28.90625,
|
| 175 |
+
"learning_rate": 9.999304101134984e-06,
|
| 176 |
+
"loss": 46.6941,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.004639327111995676,
|
| 181 |
+
"grad_norm": 30.421875,
|
| 182 |
+
"learning_rate": 9.99927510534894e-06,
|
| 183 |
+
"loss": 46.6683,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.004824900196475503,
|
| 188 |
+
"grad_norm": 29.6875,
|
| 189 |
+
"learning_rate": 9.999246109562898e-06,
|
| 190 |
+
"loss": 46.1998,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.00501047328095533,
|
| 195 |
+
"grad_norm": 27.90625,
|
| 196 |
+
"learning_rate": 9.999217113776856e-06,
|
| 197 |
+
"loss": 45.5885,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.005196046365435157,
|
| 202 |
+
"grad_norm": 31.09375,
|
| 203 |
+
"learning_rate": 9.999188117990813e-06,
|
| 204 |
+
"loss": 44.4987,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.005381619449914984,
|
| 209 |
+
"grad_norm": 32.53125,
|
| 210 |
+
"learning_rate": 9.99915912220477e-06,
|
| 211 |
+
"loss": 44.1291,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.005567192534394812,
|
| 216 |
+
"grad_norm": 27.390625,
|
| 217 |
+
"learning_rate": 9.999130126418728e-06,
|
| 218 |
+
"loss": 43.4694,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.005752765618874638,
|
| 223 |
+
"grad_norm": 29.796875,
|
| 224 |
+
"learning_rate": 9.999101130632686e-06,
|
| 225 |
+
"loss": 42.7829,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.005938338703354465,
|
| 230 |
+
"grad_norm": 33.5625,
|
| 231 |
+
"learning_rate": 9.999072134846643e-06,
|
| 232 |
+
"loss": 43.0898,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.006123911787834293,
|
| 237 |
+
"grad_norm": 28.9375,
|
| 238 |
+
"learning_rate": 9.999043139060602e-06,
|
| 239 |
+
"loss": 41.7324,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.006309484872314119,
|
| 244 |
+
"grad_norm": 29.890625,
|
| 245 |
+
"learning_rate": 9.99901414327456e-06,
|
| 246 |
+
"loss": 41.9896,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.006495057956793946,
|
| 251 |
+
"grad_norm": 35.65625,
|
| 252 |
+
"learning_rate": 9.998985147488515e-06,
|
| 253 |
+
"loss": 41.4999,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.006680631041273774,
|
| 258 |
+
"grad_norm": 28.296875,
|
| 259 |
+
"learning_rate": 9.998956151702474e-06,
|
| 260 |
+
"loss": 41.3521,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.006866204125753601,
|
| 265 |
+
"grad_norm": 27.921875,
|
| 266 |
+
"learning_rate": 9.998927155916432e-06,
|
| 267 |
+
"loss": 41.2431,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.007051777210233427,
|
| 272 |
+
"grad_norm": 34.28125,
|
| 273 |
+
"learning_rate": 9.998898160130389e-06,
|
| 274 |
+
"loss": 40.6052,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.007237350294713255,
|
| 279 |
+
"grad_norm": 27.5625,
|
| 280 |
+
"learning_rate": 9.998869164344346e-06,
|
| 281 |
+
"loss": 40.2853,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.007422923379193082,
|
| 286 |
+
"grad_norm": 30.5,
|
| 287 |
+
"learning_rate": 9.998840168558306e-06,
|
| 288 |
+
"loss": 40.2161,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.007608496463672909,
|
| 293 |
+
"grad_norm": 28.078125,
|
| 294 |
+
"learning_rate": 9.998811172772261e-06,
|
| 295 |
+
"loss": 40.0517,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.007794069548152736,
|
| 300 |
+
"grad_norm": 31.015625,
|
| 301 |
+
"learning_rate": 9.998782176986219e-06,
|
| 302 |
+
"loss": 40.1129,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.007979642632632563,
|
| 307 |
+
"grad_norm": 29.421875,
|
| 308 |
+
"learning_rate": 9.998753181200178e-06,
|
| 309 |
+
"loss": 39.3547,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.00816521571711239,
|
| 314 |
+
"grad_norm": 30.109375,
|
| 315 |
+
"learning_rate": 9.998724185414135e-06,
|
| 316 |
+
"loss": 39.3066,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.008350788801592218,
|
| 321 |
+
"grad_norm": 31.03125,
|
| 322 |
+
"learning_rate": 9.998695189628093e-06,
|
| 323 |
+
"loss": 38.9512,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.008536361886072045,
|
| 328 |
+
"grad_norm": 28.984375,
|
| 329 |
+
"learning_rate": 9.99866619384205e-06,
|
| 330 |
+
"loss": 38.7517,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.00872193497055187,
|
| 335 |
+
"grad_norm": 31.140625,
|
| 336 |
+
"learning_rate": 9.998637198056007e-06,
|
| 337 |
+
"loss": 38.4473,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.008907508055031698,
|
| 342 |
+
"grad_norm": 31.5,
|
| 343 |
+
"learning_rate": 9.998608202269965e-06,
|
| 344 |
+
"loss": 38.784,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.009093081139511525,
|
| 349 |
+
"grad_norm": 29.125,
|
| 350 |
+
"learning_rate": 9.998579206483922e-06,
|
| 351 |
+
"loss": 38.3422,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.009278654223991352,
|
| 356 |
+
"grad_norm": 29.234375,
|
| 357 |
+
"learning_rate": 9.99855021069788e-06,
|
| 358 |
+
"loss": 38.4766,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.00946422730847118,
|
| 363 |
+
"grad_norm": 30.078125,
|
| 364 |
+
"learning_rate": 9.998521214911837e-06,
|
| 365 |
+
"loss": 38.1635,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.009649800392951007,
|
| 370 |
+
"grad_norm": 29.109375,
|
| 371 |
+
"learning_rate": 9.998492219125794e-06,
|
| 372 |
+
"loss": 38.1851,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.009835373477430834,
|
| 377 |
+
"grad_norm": 29.765625,
|
| 378 |
+
"learning_rate": 9.998463223339754e-06,
|
| 379 |
+
"loss": 37.7056,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.01002094656191066,
|
| 384 |
+
"grad_norm": 30.0,
|
| 385 |
+
"learning_rate": 9.998434227553711e-06,
|
| 386 |
+
"loss": 38.0842,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.010206519646390487,
|
| 391 |
+
"grad_norm": 28.625,
|
| 392 |
+
"learning_rate": 9.998405231767668e-06,
|
| 393 |
+
"loss": 37.9516,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.010392092730870314,
|
| 398 |
+
"grad_norm": 31.1875,
|
| 399 |
+
"learning_rate": 9.998376235981626e-06,
|
| 400 |
+
"loss": 37.4985,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.010577665815350142,
|
| 405 |
+
"grad_norm": 29.65625,
|
| 406 |
+
"learning_rate": 9.998347240195583e-06,
|
| 407 |
+
"loss": 37.3159,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.010763238899829969,
|
| 412 |
+
"grad_norm": 29.578125,
|
| 413 |
+
"learning_rate": 9.99831824440954e-06,
|
| 414 |
+
"loss": 36.7818,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.010948811984309796,
|
| 419 |
+
"grad_norm": 29.90625,
|
| 420 |
+
"learning_rate": 9.998289248623498e-06,
|
| 421 |
+
"loss": 37.5521,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.011134385068789623,
|
| 426 |
+
"grad_norm": 27.546875,
|
| 427 |
+
"learning_rate": 9.998260252837455e-06,
|
| 428 |
+
"loss": 36.2803,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.011319958153269449,
|
| 433 |
+
"grad_norm": 28.84375,
|
| 434 |
+
"learning_rate": 9.998231257051415e-06,
|
| 435 |
+
"loss": 36.9916,
|
| 436 |
+
"step": 610
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.011505531237749276,
|
| 440 |
+
"grad_norm": 29.546875,
|
| 441 |
+
"learning_rate": 9.99820226126537e-06,
|
| 442 |
+
"loss": 36.3332,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.011691104322229104,
|
| 447 |
+
"grad_norm": 28.390625,
|
| 448 |
+
"learning_rate": 9.998173265479328e-06,
|
| 449 |
+
"loss": 36.4675,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.01187667740670893,
|
| 454 |
+
"grad_norm": 29.8125,
|
| 455 |
+
"learning_rate": 9.998144269693287e-06,
|
| 456 |
+
"loss": 36.7096,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.012062250491188758,
|
| 461 |
+
"grad_norm": 29.53125,
|
| 462 |
+
"learning_rate": 9.998115273907244e-06,
|
| 463 |
+
"loss": 36.3192,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.012247823575668585,
|
| 468 |
+
"grad_norm": 29.671875,
|
| 469 |
+
"learning_rate": 9.998086278121202e-06,
|
| 470 |
+
"loss": 35.995,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.012433396660148413,
|
| 475 |
+
"grad_norm": 29.0625,
|
| 476 |
+
"learning_rate": 9.998057282335159e-06,
|
| 477 |
+
"loss": 36.1418,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.012618969744628238,
|
| 482 |
+
"grad_norm": 31.671875,
|
| 483 |
+
"learning_rate": 9.998028286549116e-06,
|
| 484 |
+
"loss": 35.9698,
|
| 485 |
+
"step": 680
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.012804542829108066,
|
| 489 |
+
"grad_norm": 30.78125,
|
| 490 |
+
"learning_rate": 9.997999290763074e-06,
|
| 491 |
+
"loss": 35.9465,
|
| 492 |
+
"step": 690
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.012990115913587893,
|
| 496 |
+
"grad_norm": 30.125,
|
| 497 |
+
"learning_rate": 9.997970294977031e-06,
|
| 498 |
+
"loss": 36.0424,
|
| 499 |
+
"step": 700
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.01317568899806772,
|
| 503 |
+
"grad_norm": 30.359375,
|
| 504 |
+
"learning_rate": 9.99794129919099e-06,
|
| 505 |
+
"loss": 35.7538,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.013361262082547547,
|
| 510 |
+
"grad_norm": 31.40625,
|
| 511 |
+
"learning_rate": 9.997912303404948e-06,
|
| 512 |
+
"loss": 35.7376,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.013546835167027375,
|
| 517 |
+
"grad_norm": 31.890625,
|
| 518 |
+
"learning_rate": 9.997883307618903e-06,
|
| 519 |
+
"loss": 35.5567,
|
| 520 |
+
"step": 730
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.013732408251507202,
|
| 524 |
+
"grad_norm": 29.75,
|
| 525 |
+
"learning_rate": 9.997854311832862e-06,
|
| 526 |
+
"loss": 35.4276,
|
| 527 |
+
"step": 740
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.01391798133598703,
|
| 531 |
+
"grad_norm": 30.453125,
|
| 532 |
+
"learning_rate": 9.99782531604682e-06,
|
| 533 |
+
"loss": 35.4647,
|
| 534 |
+
"step": 750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.014103554420466855,
|
| 538 |
+
"grad_norm": 31.796875,
|
| 539 |
+
"learning_rate": 9.997796320260777e-06,
|
| 540 |
+
"loss": 35.292,
|
| 541 |
+
"step": 760
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.014289127504946682,
|
| 545 |
+
"grad_norm": 28.6875,
|
| 546 |
+
"learning_rate": 9.997767324474735e-06,
|
| 547 |
+
"loss": 35.523,
|
| 548 |
+
"step": 770
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.01447470058942651,
|
| 552 |
+
"grad_norm": 31.734375,
|
| 553 |
+
"learning_rate": 9.997738328688692e-06,
|
| 554 |
+
"loss": 34.9746,
|
| 555 |
+
"step": 780
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.014660273673906337,
|
| 559 |
+
"grad_norm": 30.0,
|
| 560 |
+
"learning_rate": 9.99770933290265e-06,
|
| 561 |
+
"loss": 35.2019,
|
| 562 |
+
"step": 790
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.014845846758386164,
|
| 566 |
+
"grad_norm": 29.859375,
|
| 567 |
+
"learning_rate": 9.997680337116607e-06,
|
| 568 |
+
"loss": 35.1678,
|
| 569 |
+
"step": 800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.015031419842865991,
|
| 573 |
+
"grad_norm": 30.796875,
|
| 574 |
+
"learning_rate": 9.997651341330566e-06,
|
| 575 |
+
"loss": 35.1566,
|
| 576 |
+
"step": 810
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.015216992927345819,
|
| 580 |
+
"grad_norm": 31.046875,
|
| 581 |
+
"learning_rate": 9.997622345544523e-06,
|
| 582 |
+
"loss": 34.756,
|
| 583 |
+
"step": 820
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.015402566011825644,
|
| 587 |
+
"grad_norm": 31.9375,
|
| 588 |
+
"learning_rate": 9.997593349758481e-06,
|
| 589 |
+
"loss": 35.1637,
|
| 590 |
+
"step": 830
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.015588139096305471,
|
| 594 |
+
"grad_norm": 30.640625,
|
| 595 |
+
"learning_rate": 9.997564353972438e-06,
|
| 596 |
+
"loss": 35.125,
|
| 597 |
+
"step": 840
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.0157737121807853,
|
| 601 |
+
"grad_norm": 30.0625,
|
| 602 |
+
"learning_rate": 9.997535358186396e-06,
|
| 603 |
+
"loss": 34.0675,
|
| 604 |
+
"step": 850
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.015959285265265126,
|
| 608 |
+
"grad_norm": 31.1875,
|
| 609 |
+
"learning_rate": 9.997506362400353e-06,
|
| 610 |
+
"loss": 34.6025,
|
| 611 |
+
"step": 860
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.016144858349744953,
|
| 615 |
+
"grad_norm": 32.59375,
|
| 616 |
+
"learning_rate": 9.99747736661431e-06,
|
| 617 |
+
"loss": 34.8561,
|
| 618 |
+
"step": 870
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.01633043143422478,
|
| 622 |
+
"grad_norm": 28.703125,
|
| 623 |
+
"learning_rate": 9.99744837082827e-06,
|
| 624 |
+
"loss": 34.5315,
|
| 625 |
+
"step": 880
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.016516004518704608,
|
| 629 |
+
"grad_norm": 32.125,
|
| 630 |
+
"learning_rate": 9.997419375042225e-06,
|
| 631 |
+
"loss": 34.2966,
|
| 632 |
+
"step": 890
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.016701577603184435,
|
| 636 |
+
"grad_norm": 31.96875,
|
| 637 |
+
"learning_rate": 9.997390379256183e-06,
|
| 638 |
+
"loss": 34.5866,
|
| 639 |
+
"step": 900
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.016887150687664262,
|
| 643 |
+
"grad_norm": 30.515625,
|
| 644 |
+
"learning_rate": 9.997361383470142e-06,
|
| 645 |
+
"loss": 34.3122,
|
| 646 |
+
"step": 910
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.01707272377214409,
|
| 650 |
+
"grad_norm": 30.078125,
|
| 651 |
+
"learning_rate": 9.9973323876841e-06,
|
| 652 |
+
"loss": 34.4973,
|
| 653 |
+
"step": 920
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.017258296856623917,
|
| 657 |
+
"grad_norm": 31.359375,
|
| 658 |
+
"learning_rate": 9.997303391898057e-06,
|
| 659 |
+
"loss": 34.3231,
|
| 660 |
+
"step": 930
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.01744386994110374,
|
| 664 |
+
"grad_norm": 31.046875,
|
| 665 |
+
"learning_rate": 9.997274396112014e-06,
|
| 666 |
+
"loss": 34.2656,
|
| 667 |
+
"step": 940
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.017629443025583568,
|
| 671 |
+
"grad_norm": 30.140625,
|
| 672 |
+
"learning_rate": 9.997245400325971e-06,
|
| 673 |
+
"loss": 34.1969,
|
| 674 |
+
"step": 950
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.017815016110063395,
|
| 678 |
+
"grad_norm": 31.96875,
|
| 679 |
+
"learning_rate": 9.997216404539929e-06,
|
| 680 |
+
"loss": 34.1862,
|
| 681 |
+
"step": 960
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.018000589194543223,
|
| 685 |
+
"grad_norm": 30.59375,
|
| 686 |
+
"learning_rate": 9.997187408753886e-06,
|
| 687 |
+
"loss": 33.5469,
|
| 688 |
+
"step": 970
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.01818616227902305,
|
| 692 |
+
"grad_norm": 31.65625,
|
| 693 |
+
"learning_rate": 9.997158412967845e-06,
|
| 694 |
+
"loss": 34.473,
|
| 695 |
+
"step": 980
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.018371735363502877,
|
| 699 |
+
"grad_norm": 31.328125,
|
| 700 |
+
"learning_rate": 9.997129417181803e-06,
|
| 701 |
+
"loss": 33.9839,
|
| 702 |
+
"step": 990
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.018557308447982705,
|
| 706 |
+
"grad_norm": 30.546875,
|
| 707 |
+
"learning_rate": 9.997100421395758e-06,
|
| 708 |
+
"loss": 33.7282,
|
| 709 |
+
"step": 1000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.018742881532462532,
|
| 713 |
+
"grad_norm": 32.90625,
|
| 714 |
+
"learning_rate": 9.997071425609718e-06,
|
| 715 |
+
"loss": 33.5576,
|
| 716 |
+
"step": 1010
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.01892845461694236,
|
| 720 |
+
"grad_norm": 30.8125,
|
| 721 |
+
"learning_rate": 9.997042429823675e-06,
|
| 722 |
+
"loss": 33.5456,
|
| 723 |
+
"step": 1020
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.019114027701422186,
|
| 727 |
+
"grad_norm": 30.09375,
|
| 728 |
+
"learning_rate": 9.997013434037632e-06,
|
| 729 |
+
"loss": 33.7308,
|
| 730 |
+
"step": 1030
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.019299600785902014,
|
| 734 |
+
"grad_norm": 30.75,
|
| 735 |
+
"learning_rate": 9.99698443825159e-06,
|
| 736 |
+
"loss": 33.1215,
|
| 737 |
+
"step": 1040
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.01948517387038184,
|
| 741 |
+
"grad_norm": 32.46875,
|
| 742 |
+
"learning_rate": 9.996955442465547e-06,
|
| 743 |
+
"loss": 33.6073,
|
| 744 |
+
"step": 1050
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.01967074695486167,
|
| 748 |
+
"grad_norm": 33.28125,
|
| 749 |
+
"learning_rate": 9.996926446679505e-06,
|
| 750 |
+
"loss": 33.6201,
|
| 751 |
+
"step": 1060
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.019856320039341496,
|
| 755 |
+
"grad_norm": 31.265625,
|
| 756 |
+
"learning_rate": 9.996897450893462e-06,
|
| 757 |
+
"loss": 33.4164,
|
| 758 |
+
"step": 1070
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.02004189312382132,
|
| 762 |
+
"grad_norm": 30.5625,
|
| 763 |
+
"learning_rate": 9.99686845510742e-06,
|
| 764 |
+
"loss": 33.5612,
|
| 765 |
+
"step": 1080
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.020227466208301147,
|
| 769 |
+
"grad_norm": 30.9375,
|
| 770 |
+
"learning_rate": 9.996839459321379e-06,
|
| 771 |
+
"loss": 33.128,
|
| 772 |
+
"step": 1090
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.020413039292780974,
|
| 776 |
+
"grad_norm": 30.390625,
|
| 777 |
+
"learning_rate": 9.996810463535334e-06,
|
| 778 |
+
"loss": 33.3032,
|
| 779 |
+
"step": 1100
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.0205986123772608,
|
| 783 |
+
"grad_norm": 31.8125,
|
| 784 |
+
"learning_rate": 9.996781467749292e-06,
|
| 785 |
+
"loss": 32.6613,
|
| 786 |
+
"step": 1110
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.02078418546174063,
|
| 790 |
+
"grad_norm": 31.5,
|
| 791 |
+
"learning_rate": 9.99675247196325e-06,
|
| 792 |
+
"loss": 32.9665,
|
| 793 |
+
"step": 1120
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.020969758546220456,
|
| 797 |
+
"grad_norm": 30.828125,
|
| 798 |
+
"learning_rate": 9.996723476177208e-06,
|
| 799 |
+
"loss": 33.3112,
|
| 800 |
+
"step": 1130
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.021155331630700283,
|
| 804 |
+
"grad_norm": 33.0,
|
| 805 |
+
"learning_rate": 9.996694480391166e-06,
|
| 806 |
+
"loss": 33.3407,
|
| 807 |
+
"step": 1140
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.02134090471518011,
|
| 811 |
+
"grad_norm": 32.34375,
|
| 812 |
+
"learning_rate": 9.996665484605123e-06,
|
| 813 |
+
"loss": 33.1562,
|
| 814 |
+
"step": 1150
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.021526477799659938,
|
| 818 |
+
"grad_norm": 31.96875,
|
| 819 |
+
"learning_rate": 9.99663648881908e-06,
|
| 820 |
+
"loss": 32.8546,
|
| 821 |
+
"step": 1160
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.021712050884139765,
|
| 825 |
+
"grad_norm": 32.28125,
|
| 826 |
+
"learning_rate": 9.996607493033038e-06,
|
| 827 |
+
"loss": 32.8359,
|
| 828 |
+
"step": 1170
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.021897623968619592,
|
| 832 |
+
"grad_norm": 31.484375,
|
| 833 |
+
"learning_rate": 9.996578497246995e-06,
|
| 834 |
+
"loss": 32.6904,
|
| 835 |
+
"step": 1180
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.02208319705309942,
|
| 839 |
+
"grad_norm": 33.9375,
|
| 840 |
+
"learning_rate": 9.996549501460954e-06,
|
| 841 |
+
"loss": 33.2044,
|
| 842 |
+
"step": 1190
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.022268770137579247,
|
| 846 |
+
"grad_norm": 33.03125,
|
| 847 |
+
"learning_rate": 9.996520505674912e-06,
|
| 848 |
+
"loss": 33.0732,
|
| 849 |
+
"step": 1200
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.022454343222059074,
|
| 853 |
+
"grad_norm": 30.65625,
|
| 854 |
+
"learning_rate": 9.996491509888867e-06,
|
| 855 |
+
"loss": 32.7397,
|
| 856 |
+
"step": 1210
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.022639916306538898,
|
| 860 |
+
"grad_norm": 32.5625,
|
| 861 |
+
"learning_rate": 9.996462514102827e-06,
|
| 862 |
+
"loss": 32.2421,
|
| 863 |
+
"step": 1220
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.022825489391018725,
|
| 867 |
+
"grad_norm": 31.390625,
|
| 868 |
+
"learning_rate": 9.996433518316784e-06,
|
| 869 |
+
"loss": 32.6799,
|
| 870 |
+
"step": 1230
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.023011062475498553,
|
| 874 |
+
"grad_norm": 32.625,
|
| 875 |
+
"learning_rate": 9.996404522530741e-06,
|
| 876 |
+
"loss": 32.6819,
|
| 877 |
+
"step": 1240
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.02319663555997838,
|
| 881 |
+
"grad_norm": 31.046875,
|
| 882 |
+
"learning_rate": 9.996375526744699e-06,
|
| 883 |
+
"loss": 32.3593,
|
| 884 |
+
"step": 1250
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.023382208644458207,
|
| 888 |
+
"grad_norm": 31.96875,
|
| 889 |
+
"learning_rate": 9.996346530958658e-06,
|
| 890 |
+
"loss": 32.3033,
|
| 891 |
+
"step": 1260
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.023567781728938034,
|
| 895 |
+
"grad_norm": 31.71875,
|
| 896 |
+
"learning_rate": 9.996317535172614e-06,
|
| 897 |
+
"loss": 32.0812,
|
| 898 |
+
"step": 1270
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.02375335481341786,
|
| 902 |
+
"grad_norm": 32.75,
|
| 903 |
+
"learning_rate": 9.996288539386571e-06,
|
| 904 |
+
"loss": 32.636,
|
| 905 |
+
"step": 1280
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.02393892789789769,
|
| 909 |
+
"grad_norm": 30.765625,
|
| 910 |
+
"learning_rate": 9.99625954360053e-06,
|
| 911 |
+
"loss": 32.2014,
|
| 912 |
+
"step": 1290
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.024124500982377516,
|
| 916 |
+
"grad_norm": 32.625,
|
| 917 |
+
"learning_rate": 9.996230547814487e-06,
|
| 918 |
+
"loss": 32.5447,
|
| 919 |
+
"step": 1300
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.024310074066857344,
|
| 923 |
+
"grad_norm": 32.5,
|
| 924 |
+
"learning_rate": 9.996201552028445e-06,
|
| 925 |
+
"loss": 32.4539,
|
| 926 |
+
"step": 1310
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.02449564715133717,
|
| 930 |
+
"grad_norm": 30.59375,
|
| 931 |
+
"learning_rate": 9.996172556242402e-06,
|
| 932 |
+
"loss": 32.3299,
|
| 933 |
+
"step": 1320
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.024681220235816998,
|
| 937 |
+
"grad_norm": 31.46875,
|
| 938 |
+
"learning_rate": 9.99614356045636e-06,
|
| 939 |
+
"loss": 32.2265,
|
| 940 |
+
"step": 1330
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.024866793320296825,
|
| 944 |
+
"grad_norm": 32.65625,
|
| 945 |
+
"learning_rate": 9.996114564670317e-06,
|
| 946 |
+
"loss": 32.1521,
|
| 947 |
+
"step": 1340
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.025052366404776653,
|
| 951 |
+
"grad_norm": 32.0625,
|
| 952 |
+
"learning_rate": 9.996085568884274e-06,
|
| 953 |
+
"loss": 32.5649,
|
| 954 |
+
"step": 1350
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.025237939489256477,
|
| 958 |
+
"grad_norm": 32.6875,
|
| 959 |
+
"learning_rate": 9.996056573098234e-06,
|
| 960 |
+
"loss": 32.2027,
|
| 961 |
+
"step": 1360
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.025423512573736304,
|
| 965 |
+
"grad_norm": 32.28125,
|
| 966 |
+
"learning_rate": 9.99602757731219e-06,
|
| 967 |
+
"loss": 32.4088,
|
| 968 |
+
"step": 1370
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.02560908565821613,
|
| 972 |
+
"grad_norm": 31.21875,
|
| 973 |
+
"learning_rate": 9.995998581526147e-06,
|
| 974 |
+
"loss": 32.554,
|
| 975 |
+
"step": 1380
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.02579465874269596,
|
| 979 |
+
"grad_norm": 33.5,
|
| 980 |
+
"learning_rate": 9.995969585740106e-06,
|
| 981 |
+
"loss": 32.3441,
|
| 982 |
+
"step": 1390
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.025980231827175786,
|
| 986 |
+
"grad_norm": 31.484375,
|
| 987 |
+
"learning_rate": 9.995940589954063e-06,
|
| 988 |
+
"loss": 31.8443,
|
| 989 |
+
"step": 1400
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.026165804911655613,
|
| 993 |
+
"grad_norm": 32.25,
|
| 994 |
+
"learning_rate": 9.99591159416802e-06,
|
| 995 |
+
"loss": 32.0743,
|
| 996 |
+
"step": 1410
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.02635137799613544,
|
| 1000 |
+
"grad_norm": 32.03125,
|
| 1001 |
+
"learning_rate": 9.995882598381978e-06,
|
| 1002 |
+
"loss": 31.8419,
|
| 1003 |
+
"step": 1420
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.026536951080615268,
|
| 1007 |
+
"grad_norm": 33.625,
|
| 1008 |
+
"learning_rate": 9.995853602595935e-06,
|
| 1009 |
+
"loss": 31.8767,
|
| 1010 |
+
"step": 1430
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.026722524165095095,
|
| 1014 |
+
"grad_norm": 30.9375,
|
| 1015 |
+
"learning_rate": 9.995824606809893e-06,
|
| 1016 |
+
"loss": 31.7176,
|
| 1017 |
+
"step": 1440
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.026908097249574922,
|
| 1021 |
+
"grad_norm": 31.234375,
|
| 1022 |
+
"learning_rate": 9.99579561102385e-06,
|
| 1023 |
+
"loss": 32.3033,
|
| 1024 |
+
"step": 1450
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.02709367033405475,
|
| 1028 |
+
"grad_norm": 30.828125,
|
| 1029 |
+
"learning_rate": 9.99576661523781e-06,
|
| 1030 |
+
"loss": 31.5171,
|
| 1031 |
+
"step": 1460
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.027279243418534577,
|
| 1035 |
+
"grad_norm": 30.0625,
|
| 1036 |
+
"learning_rate": 9.995737619451767e-06,
|
| 1037 |
+
"loss": 31.8232,
|
| 1038 |
+
"step": 1470
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.027464816503014404,
|
| 1042 |
+
"grad_norm": 33.34375,
|
| 1043 |
+
"learning_rate": 9.995708623665722e-06,
|
| 1044 |
+
"loss": 31.6173,
|
| 1045 |
+
"step": 1480
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.02765038958749423,
|
| 1049 |
+
"grad_norm": 32.09375,
|
| 1050 |
+
"learning_rate": 9.995679627879682e-06,
|
| 1051 |
+
"loss": 31.9755,
|
| 1052 |
+
"step": 1490
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.02783596267197406,
|
| 1056 |
+
"grad_norm": 33.03125,
|
| 1057 |
+
"learning_rate": 9.995650632093639e-06,
|
| 1058 |
+
"loss": 32.1813,
|
| 1059 |
+
"step": 1500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.028021535756453882,
|
| 1063 |
+
"grad_norm": 30.984375,
|
| 1064 |
+
"learning_rate": 9.995621636307596e-06,
|
| 1065 |
+
"loss": 31.64,
|
| 1066 |
+
"step": 1510
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.02820710884093371,
|
| 1070 |
+
"grad_norm": 31.828125,
|
| 1071 |
+
"learning_rate": 9.995592640521554e-06,
|
| 1072 |
+
"loss": 31.9134,
|
| 1073 |
+
"step": 1520
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.028392681925413537,
|
| 1077 |
+
"grad_norm": 33.28125,
|
| 1078 |
+
"learning_rate": 9.995563644735511e-06,
|
| 1079 |
+
"loss": 31.2862,
|
| 1080 |
+
"step": 1530
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.028578255009893364,
|
| 1084 |
+
"grad_norm": 31.984375,
|
| 1085 |
+
"learning_rate": 9.995534648949469e-06,
|
| 1086 |
+
"loss": 31.092,
|
| 1087 |
+
"step": 1540
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.02876382809437319,
|
| 1091 |
+
"grad_norm": 31.015625,
|
| 1092 |
+
"learning_rate": 9.995505653163426e-06,
|
| 1093 |
+
"loss": 31.5553,
|
| 1094 |
+
"step": 1550
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.02894940117885302,
|
| 1098 |
+
"grad_norm": 32.1875,
|
| 1099 |
+
"learning_rate": 9.995476657377383e-06,
|
| 1100 |
+
"loss": 31.8129,
|
| 1101 |
+
"step": 1560
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.029134974263332846,
|
| 1105 |
+
"grad_norm": 31.859375,
|
| 1106 |
+
"learning_rate": 9.995447661591343e-06,
|
| 1107 |
+
"loss": 31.3235,
|
| 1108 |
+
"step": 1570
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.029320547347812673,
|
| 1112 |
+
"grad_norm": 31.984375,
|
| 1113 |
+
"learning_rate": 9.9954186658053e-06,
|
| 1114 |
+
"loss": 31.2428,
|
| 1115 |
+
"step": 1580
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.0295061204322925,
|
| 1119 |
+
"grad_norm": 32.8125,
|
| 1120 |
+
"learning_rate": 9.995389670019257e-06,
|
| 1121 |
+
"loss": 31.4959,
|
| 1122 |
+
"step": 1590
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.029691693516772328,
|
| 1126 |
+
"grad_norm": 32.65625,
|
| 1127 |
+
"learning_rate": 9.995360674233215e-06,
|
| 1128 |
+
"loss": 31.7104,
|
| 1129 |
+
"step": 1600
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.029877266601252155,
|
| 1133 |
+
"grad_norm": 31.828125,
|
| 1134 |
+
"learning_rate": 9.995331678447172e-06,
|
| 1135 |
+
"loss": 31.4195,
|
| 1136 |
+
"step": 1610
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.030062839685731983,
|
| 1140 |
+
"grad_norm": 31.4375,
|
| 1141 |
+
"learning_rate": 9.99530268266113e-06,
|
| 1142 |
+
"loss": 31.5362,
|
| 1143 |
+
"step": 1620
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.03024841277021181,
|
| 1147 |
+
"grad_norm": 30.09375,
|
| 1148 |
+
"learning_rate": 9.995273686875087e-06,
|
| 1149 |
+
"loss": 31.0253,
|
| 1150 |
+
"step": 1630
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.030433985854691637,
|
| 1154 |
+
"grad_norm": 33.53125,
|
| 1155 |
+
"learning_rate": 9.995244691089044e-06,
|
| 1156 |
+
"loss": 31.1842,
|
| 1157 |
+
"step": 1640
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.03061955893917146,
|
| 1161 |
+
"grad_norm": 30.609375,
|
| 1162 |
+
"learning_rate": 9.995215695303002e-06,
|
| 1163 |
+
"loss": 31.149,
|
| 1164 |
+
"step": 1650
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.030805132023651288,
|
| 1168 |
+
"grad_norm": 31.71875,
|
| 1169 |
+
"learning_rate": 9.99518669951696e-06,
|
| 1170 |
+
"loss": 31.649,
|
| 1171 |
+
"step": 1660
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.030990705108131116,
|
| 1175 |
+
"grad_norm": 31.53125,
|
| 1176 |
+
"learning_rate": 9.995157703730918e-06,
|
| 1177 |
+
"loss": 30.8072,
|
| 1178 |
+
"step": 1670
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.031176278192610943,
|
| 1182 |
+
"grad_norm": 32.25,
|
| 1183 |
+
"learning_rate": 9.995128707944876e-06,
|
| 1184 |
+
"loss": 31.2502,
|
| 1185 |
+
"step": 1680
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.03136185127709077,
|
| 1189 |
+
"grad_norm": 31.140625,
|
| 1190 |
+
"learning_rate": 9.995099712158831e-06,
|
| 1191 |
+
"loss": 31.1485,
|
| 1192 |
+
"step": 1690
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.0315474243615706,
|
| 1196 |
+
"grad_norm": 33.125,
|
| 1197 |
+
"learning_rate": 9.99507071637279e-06,
|
| 1198 |
+
"loss": 31.5347,
|
| 1199 |
+
"step": 1700
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.031732997446050425,
|
| 1203 |
+
"grad_norm": 30.890625,
|
| 1204 |
+
"learning_rate": 9.995041720586748e-06,
|
| 1205 |
+
"loss": 31.1346,
|
| 1206 |
+
"step": 1710
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.03191857053053025,
|
| 1210 |
+
"grad_norm": 30.640625,
|
| 1211 |
+
"learning_rate": 9.995012724800705e-06,
|
| 1212 |
+
"loss": 31.0724,
|
| 1213 |
+
"step": 1720
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.03210414361501008,
|
| 1217 |
+
"grad_norm": 29.265625,
|
| 1218 |
+
"learning_rate": 9.994983729014663e-06,
|
| 1219 |
+
"loss": 31.0588,
|
| 1220 |
+
"step": 1730
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.032289716699489907,
|
| 1224 |
+
"grad_norm": 33.53125,
|
| 1225 |
+
"learning_rate": 9.994954733228622e-06,
|
| 1226 |
+
"loss": 31.6928,
|
| 1227 |
+
"step": 1740
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.032475289783969734,
|
| 1231 |
+
"grad_norm": 33.75,
|
| 1232 |
+
"learning_rate": 9.994925737442578e-06,
|
| 1233 |
+
"loss": 31.1413,
|
| 1234 |
+
"step": 1750
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.03266086286844956,
|
| 1238 |
+
"grad_norm": 33.125,
|
| 1239 |
+
"learning_rate": 9.994896741656535e-06,
|
| 1240 |
+
"loss": 30.7202,
|
| 1241 |
+
"step": 1760
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.03284643595292939,
|
| 1245 |
+
"grad_norm": 33.78125,
|
| 1246 |
+
"learning_rate": 9.994867745870494e-06,
|
| 1247 |
+
"loss": 30.409,
|
| 1248 |
+
"step": 1770
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.033032009037409216,
|
| 1252 |
+
"grad_norm": 33.15625,
|
| 1253 |
+
"learning_rate": 9.994838750084451e-06,
|
| 1254 |
+
"loss": 31.4688,
|
| 1255 |
+
"step": 1780
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.03321758212188904,
|
| 1259 |
+
"grad_norm": 31.953125,
|
| 1260 |
+
"learning_rate": 9.994809754298409e-06,
|
| 1261 |
+
"loss": 30.6562,
|
| 1262 |
+
"step": 1790
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.03340315520636887,
|
| 1266 |
+
"grad_norm": 33.25,
|
| 1267 |
+
"learning_rate": 9.994780758512366e-06,
|
| 1268 |
+
"loss": 30.7922,
|
| 1269 |
+
"step": 1800
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.0335887282908487,
|
| 1273 |
+
"grad_norm": 32.96875,
|
| 1274 |
+
"learning_rate": 9.994751762726324e-06,
|
| 1275 |
+
"loss": 30.7261,
|
| 1276 |
+
"step": 1810
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.033774301375328525,
|
| 1280 |
+
"grad_norm": 30.546875,
|
| 1281 |
+
"learning_rate": 9.994722766940281e-06,
|
| 1282 |
+
"loss": 30.6429,
|
| 1283 |
+
"step": 1820
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.03395987445980835,
|
| 1287 |
+
"grad_norm": 31.796875,
|
| 1288 |
+
"learning_rate": 9.994693771154239e-06,
|
| 1289 |
+
"loss": 31.0909,
|
| 1290 |
+
"step": 1830
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.03414544754428818,
|
| 1294 |
+
"grad_norm": 30.84375,
|
| 1295 |
+
"learning_rate": 9.994664775368198e-06,
|
| 1296 |
+
"loss": 30.5669,
|
| 1297 |
+
"step": 1840
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.03433102062876801,
|
| 1301 |
+
"grad_norm": 31.65625,
|
| 1302 |
+
"learning_rate": 9.994635779582155e-06,
|
| 1303 |
+
"loss": 30.707,
|
| 1304 |
+
"step": 1850
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.034516593713247834,
|
| 1308 |
+
"grad_norm": 31.1875,
|
| 1309 |
+
"learning_rate": 9.99460678379611e-06,
|
| 1310 |
+
"loss": 31.0501,
|
| 1311 |
+
"step": 1860
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.034702166797727654,
|
| 1315 |
+
"grad_norm": 33.5,
|
| 1316 |
+
"learning_rate": 9.99457778801007e-06,
|
| 1317 |
+
"loss": 30.9713,
|
| 1318 |
+
"step": 1870
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.03488773988220748,
|
| 1322 |
+
"grad_norm": 33.84375,
|
| 1323 |
+
"learning_rate": 9.994548792224027e-06,
|
| 1324 |
+
"loss": 30.8638,
|
| 1325 |
+
"step": 1880
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.03507331296668731,
|
| 1329 |
+
"grad_norm": 32.5,
|
| 1330 |
+
"learning_rate": 9.994519796437985e-06,
|
| 1331 |
+
"loss": 31.0086,
|
| 1332 |
+
"step": 1890
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.035258886051167136,
|
| 1336 |
+
"grad_norm": 33.8125,
|
| 1337 |
+
"learning_rate": 9.994490800651942e-06,
|
| 1338 |
+
"loss": 30.5325,
|
| 1339 |
+
"step": 1900
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.035444459135646963,
|
| 1343 |
+
"grad_norm": 33.4375,
|
| 1344 |
+
"learning_rate": 9.9944618048659e-06,
|
| 1345 |
+
"loss": 30.611,
|
| 1346 |
+
"step": 1910
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.03563003222012679,
|
| 1350 |
+
"grad_norm": 31.984375,
|
| 1351 |
+
"learning_rate": 9.994432809079857e-06,
|
| 1352 |
+
"loss": 30.3778,
|
| 1353 |
+
"step": 1920
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.03581560530460662,
|
| 1357 |
+
"grad_norm": 32.40625,
|
| 1358 |
+
"learning_rate": 9.994403813293814e-06,
|
| 1359 |
+
"loss": 30.6509,
|
| 1360 |
+
"step": 1930
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.036001178389086445,
|
| 1364 |
+
"grad_norm": 31.796875,
|
| 1365 |
+
"learning_rate": 9.994374817507773e-06,
|
| 1366 |
+
"loss": 30.8435,
|
| 1367 |
+
"step": 1940
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.03618675147356627,
|
| 1371 |
+
"grad_norm": 32.75,
|
| 1372 |
+
"learning_rate": 9.99434582172173e-06,
|
| 1373 |
+
"loss": 30.1924,
|
| 1374 |
+
"step": 1950
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.0363723245580461,
|
| 1378 |
+
"grad_norm": 31.984375,
|
| 1379 |
+
"learning_rate": 9.994316825935686e-06,
|
| 1380 |
+
"loss": 30.4882,
|
| 1381 |
+
"step": 1960
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.03655789764252593,
|
| 1385 |
+
"grad_norm": 33.40625,
|
| 1386 |
+
"learning_rate": 9.994287830149646e-06,
|
| 1387 |
+
"loss": 30.4694,
|
| 1388 |
+
"step": 1970
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.036743470727005755,
|
| 1392 |
+
"grad_norm": 33.96875,
|
| 1393 |
+
"learning_rate": 9.994258834363603e-06,
|
| 1394 |
+
"loss": 30.297,
|
| 1395 |
+
"step": 1980
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.03692904381148558,
|
| 1399 |
+
"grad_norm": 32.71875,
|
| 1400 |
+
"learning_rate": 9.99422983857756e-06,
|
| 1401 |
+
"loss": 30.2815,
|
| 1402 |
+
"step": 1990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.03711461689596541,
|
| 1406 |
+
"grad_norm": 31.96875,
|
| 1407 |
+
"learning_rate": 9.994200842791518e-06,
|
| 1408 |
+
"loss": 30.7292,
|
| 1409 |
+
"step": 2000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.037300189980445236,
|
| 1413 |
+
"grad_norm": 32.96875,
|
| 1414 |
+
"learning_rate": 9.994171847005475e-06,
|
| 1415 |
+
"loss": 30.7787,
|
| 1416 |
+
"step": 2010
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.037485763064925064,
|
| 1420 |
+
"grad_norm": 33.59375,
|
| 1421 |
+
"learning_rate": 9.994142851219433e-06,
|
| 1422 |
+
"loss": 30.5994,
|
| 1423 |
+
"step": 2020
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.03767133614940489,
|
| 1427 |
+
"grad_norm": 32.5,
|
| 1428 |
+
"learning_rate": 9.99411385543339e-06,
|
| 1429 |
+
"loss": 30.2407,
|
| 1430 |
+
"step": 2030
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.03785690923388472,
|
| 1434 |
+
"grad_norm": 30.28125,
|
| 1435 |
+
"learning_rate": 9.994084859647349e-06,
|
| 1436 |
+
"loss": 29.9962,
|
| 1437 |
+
"step": 2040
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.038042482318364546,
|
| 1441 |
+
"grad_norm": 36.53125,
|
| 1442 |
+
"learning_rate": 9.994055863861307e-06,
|
| 1443 |
+
"loss": 30.1751,
|
| 1444 |
+
"step": 2050
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.03822805540284437,
|
| 1448 |
+
"grad_norm": 34.3125,
|
| 1449 |
+
"learning_rate": 9.994026868075264e-06,
|
| 1450 |
+
"loss": 30.1296,
|
| 1451 |
+
"step": 2060
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.0384136284873242,
|
| 1455 |
+
"grad_norm": 31.78125,
|
| 1456 |
+
"learning_rate": 9.993997872289221e-06,
|
| 1457 |
+
"loss": 29.8851,
|
| 1458 |
+
"step": 2070
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.03859920157180403,
|
| 1462 |
+
"grad_norm": 34.3125,
|
| 1463 |
+
"learning_rate": 9.993968876503179e-06,
|
| 1464 |
+
"loss": 30.1799,
|
| 1465 |
+
"step": 2080
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.038784774656283855,
|
| 1469 |
+
"grad_norm": 32.5625,
|
| 1470 |
+
"learning_rate": 9.993939880717136e-06,
|
| 1471 |
+
"loss": 30.1617,
|
| 1472 |
+
"step": 2090
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.03897034774076368,
|
| 1476 |
+
"grad_norm": 31.671875,
|
| 1477 |
+
"learning_rate": 9.993910884931094e-06,
|
| 1478 |
+
"loss": 30.4521,
|
| 1479 |
+
"step": 2100
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.03915592082524351,
|
| 1483 |
+
"grad_norm": 33.34375,
|
| 1484 |
+
"learning_rate": 9.993881889145051e-06,
|
| 1485 |
+
"loss": 30.1233,
|
| 1486 |
+
"step": 2110
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.03934149390972334,
|
| 1490 |
+
"grad_norm": 32.65625,
|
| 1491 |
+
"learning_rate": 9.993852893359008e-06,
|
| 1492 |
+
"loss": 30.265,
|
| 1493 |
+
"step": 2120
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.039527066994203164,
|
| 1497 |
+
"grad_norm": 31.859375,
|
| 1498 |
+
"learning_rate": 9.993823897572966e-06,
|
| 1499 |
+
"loss": 29.874,
|
| 1500 |
+
"step": 2130
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.03971264007868299,
|
| 1504 |
+
"grad_norm": 32.8125,
|
| 1505 |
+
"learning_rate": 9.993794901786923e-06,
|
| 1506 |
+
"loss": 30.2309,
|
| 1507 |
+
"step": 2140
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.03989821316316281,
|
| 1511 |
+
"grad_norm": 33.0,
|
| 1512 |
+
"learning_rate": 9.993765906000882e-06,
|
| 1513 |
+
"loss": 30.0726,
|
| 1514 |
+
"step": 2150
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.04008378624764264,
|
| 1518 |
+
"grad_norm": 33.0,
|
| 1519 |
+
"learning_rate": 9.99373691021484e-06,
|
| 1520 |
+
"loss": 30.0015,
|
| 1521 |
+
"step": 2160
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.040269359332122466,
|
| 1525 |
+
"grad_norm": 32.96875,
|
| 1526 |
+
"learning_rate": 9.993707914428797e-06,
|
| 1527 |
+
"loss": 30.4088,
|
| 1528 |
+
"step": 2170
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.04045493241660229,
|
| 1532 |
+
"grad_norm": 32.21875,
|
| 1533 |
+
"learning_rate": 9.993678918642755e-06,
|
| 1534 |
+
"loss": 29.5758,
|
| 1535 |
+
"step": 2180
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.04064050550108212,
|
| 1539 |
+
"grad_norm": 32.6875,
|
| 1540 |
+
"learning_rate": 9.993649922856712e-06,
|
| 1541 |
+
"loss": 30.0286,
|
| 1542 |
+
"step": 2190
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.04082607858556195,
|
| 1546 |
+
"grad_norm": 31.515625,
|
| 1547 |
+
"learning_rate": 9.99362092707067e-06,
|
| 1548 |
+
"loss": 29.7122,
|
| 1549 |
+
"step": 2200
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.041011651670041775,
|
| 1553 |
+
"grad_norm": 31.375,
|
| 1554 |
+
"learning_rate": 9.993591931284627e-06,
|
| 1555 |
+
"loss": 30.272,
|
| 1556 |
+
"step": 2210
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.0411972247545216,
|
| 1560 |
+
"grad_norm": 32.84375,
|
| 1561 |
+
"learning_rate": 9.993562935498586e-06,
|
| 1562 |
+
"loss": 29.549,
|
| 1563 |
+
"step": 2220
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.04138279783900143,
|
| 1567 |
+
"grad_norm": 32.5625,
|
| 1568 |
+
"learning_rate": 9.993533939712542e-06,
|
| 1569 |
+
"loss": 29.9526,
|
| 1570 |
+
"step": 2230
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.04156837092348126,
|
| 1574 |
+
"grad_norm": 32.75,
|
| 1575 |
+
"learning_rate": 9.993504943926499e-06,
|
| 1576 |
+
"loss": 29.7783,
|
| 1577 |
+
"step": 2240
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.041753944007961084,
|
| 1581 |
+
"grad_norm": 33.375,
|
| 1582 |
+
"learning_rate": 9.993475948140458e-06,
|
| 1583 |
+
"loss": 29.815,
|
| 1584 |
+
"step": 2250
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.04193951709244091,
|
| 1588 |
+
"grad_norm": 31.921875,
|
| 1589 |
+
"learning_rate": 9.993446952354415e-06,
|
| 1590 |
+
"loss": 30.2447,
|
| 1591 |
+
"step": 2260
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.04212509017692074,
|
| 1595 |
+
"grad_norm": 33.71875,
|
| 1596 |
+
"learning_rate": 9.993417956568373e-06,
|
| 1597 |
+
"loss": 30.0316,
|
| 1598 |
+
"step": 2270
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.042310663261400566,
|
| 1602 |
+
"grad_norm": 31.609375,
|
| 1603 |
+
"learning_rate": 9.99338896078233e-06,
|
| 1604 |
+
"loss": 29.8423,
|
| 1605 |
+
"step": 2280
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.042496236345880394,
|
| 1609 |
+
"grad_norm": 30.828125,
|
| 1610 |
+
"learning_rate": 9.993359964996288e-06,
|
| 1611 |
+
"loss": 29.8542,
|
| 1612 |
+
"step": 2290
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.04268180943036022,
|
| 1616 |
+
"grad_norm": 32.15625,
|
| 1617 |
+
"learning_rate": 9.993330969210245e-06,
|
| 1618 |
+
"loss": 29.8611,
|
| 1619 |
+
"step": 2300
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.04286738251484005,
|
| 1623 |
+
"grad_norm": 33.03125,
|
| 1624 |
+
"learning_rate": 9.993301973424203e-06,
|
| 1625 |
+
"loss": 29.6654,
|
| 1626 |
+
"step": 2310
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.043052955599319875,
|
| 1630 |
+
"grad_norm": 33.6875,
|
| 1631 |
+
"learning_rate": 9.993272977638162e-06,
|
| 1632 |
+
"loss": 29.5965,
|
| 1633 |
+
"step": 2320
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.0432385286837997,
|
| 1637 |
+
"grad_norm": 33.5625,
|
| 1638 |
+
"learning_rate": 9.993243981852119e-06,
|
| 1639 |
+
"loss": 29.4433,
|
| 1640 |
+
"step": 2330
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.04342410176827953,
|
| 1644 |
+
"grad_norm": 32.71875,
|
| 1645 |
+
"learning_rate": 9.993214986066075e-06,
|
| 1646 |
+
"loss": 29.2134,
|
| 1647 |
+
"step": 2340
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.04360967485275936,
|
| 1651 |
+
"grad_norm": 33.09375,
|
| 1652 |
+
"learning_rate": 9.993185990280034e-06,
|
| 1653 |
+
"loss": 29.5822,
|
| 1654 |
+
"step": 2350
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.043795247937239185,
|
| 1658 |
+
"grad_norm": 34.5625,
|
| 1659 |
+
"learning_rate": 9.993156994493991e-06,
|
| 1660 |
+
"loss": 29.5265,
|
| 1661 |
+
"step": 2360
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.04398082102171901,
|
| 1665 |
+
"grad_norm": 32.8125,
|
| 1666 |
+
"learning_rate": 9.993127998707949e-06,
|
| 1667 |
+
"loss": 29.9218,
|
| 1668 |
+
"step": 2370
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.04416639410619884,
|
| 1672 |
+
"grad_norm": 34.25,
|
| 1673 |
+
"learning_rate": 9.993099002921906e-06,
|
| 1674 |
+
"loss": 29.5988,
|
| 1675 |
+
"step": 2380
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.044351967190678666,
|
| 1679 |
+
"grad_norm": 32.84375,
|
| 1680 |
+
"learning_rate": 9.993070007135863e-06,
|
| 1681 |
+
"loss": 29.2194,
|
| 1682 |
+
"step": 2390
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.044537540275158494,
|
| 1686 |
+
"grad_norm": 33.25,
|
| 1687 |
+
"learning_rate": 9.993041011349821e-06,
|
| 1688 |
+
"loss": 29.3336,
|
| 1689 |
+
"step": 2400
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.04472311335963832,
|
| 1693 |
+
"grad_norm": 32.625,
|
| 1694 |
+
"learning_rate": 9.993012015563778e-06,
|
| 1695 |
+
"loss": 29.1395,
|
| 1696 |
+
"step": 2410
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.04490868644411815,
|
| 1700 |
+
"grad_norm": 31.890625,
|
| 1701 |
+
"learning_rate": 9.992983019777737e-06,
|
| 1702 |
+
"loss": 29.6973,
|
| 1703 |
+
"step": 2420
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.045094259528597976,
|
| 1707 |
+
"grad_norm": 35.0625,
|
| 1708 |
+
"learning_rate": 9.992954023991695e-06,
|
| 1709 |
+
"loss": 30.2094,
|
| 1710 |
+
"step": 2430
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.045279832613077796,
|
| 1714 |
+
"grad_norm": 30.609375,
|
| 1715 |
+
"learning_rate": 9.992925028205652e-06,
|
| 1716 |
+
"loss": 29.4222,
|
| 1717 |
+
"step": 2440
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.04546540569755762,
|
| 1721 |
+
"grad_norm": 34.1875,
|
| 1722 |
+
"learning_rate": 9.99289603241961e-06,
|
| 1723 |
+
"loss": 29.5045,
|
| 1724 |
+
"step": 2450
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.04565097878203745,
|
| 1728 |
+
"grad_norm": 33.15625,
|
| 1729 |
+
"learning_rate": 9.992867036633567e-06,
|
| 1730 |
+
"loss": 29.2573,
|
| 1731 |
+
"step": 2460
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.04583655186651728,
|
| 1735 |
+
"grad_norm": 33.34375,
|
| 1736 |
+
"learning_rate": 9.992838040847524e-06,
|
| 1737 |
+
"loss": 29.4728,
|
| 1738 |
+
"step": 2470
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.046022124950997105,
|
| 1742 |
+
"grad_norm": 36.40625,
|
| 1743 |
+
"learning_rate": 9.992809045061482e-06,
|
| 1744 |
+
"loss": 29.705,
|
| 1745 |
+
"step": 2480
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.04620769803547693,
|
| 1749 |
+
"grad_norm": 32.65625,
|
| 1750 |
+
"learning_rate": 9.992780049275441e-06,
|
| 1751 |
+
"loss": 29.8484,
|
| 1752 |
+
"step": 2490
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.04639327111995676,
|
| 1756 |
+
"grad_norm": 34.75,
|
| 1757 |
+
"learning_rate": 9.992751053489397e-06,
|
| 1758 |
+
"loss": 29.621,
|
| 1759 |
+
"step": 2500
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.04657884420443659,
|
| 1763 |
+
"grad_norm": 33.71875,
|
| 1764 |
+
"learning_rate": 9.992722057703354e-06,
|
| 1765 |
+
"loss": 29.9103,
|
| 1766 |
+
"step": 2510
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.046764417288916414,
|
| 1770 |
+
"grad_norm": 32.53125,
|
| 1771 |
+
"learning_rate": 9.992693061917313e-06,
|
| 1772 |
+
"loss": 29.0736,
|
| 1773 |
+
"step": 2520
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.04694999037339624,
|
| 1777 |
+
"grad_norm": 31.203125,
|
| 1778 |
+
"learning_rate": 9.99266406613127e-06,
|
| 1779 |
+
"loss": 29.5914,
|
| 1780 |
+
"step": 2530
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.04713556345787607,
|
| 1784 |
+
"grad_norm": 32.125,
|
| 1785 |
+
"learning_rate": 9.992635070345228e-06,
|
| 1786 |
+
"loss": 28.9369,
|
| 1787 |
+
"step": 2540
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.047321136542355896,
|
| 1791 |
+
"grad_norm": 32.28125,
|
| 1792 |
+
"learning_rate": 9.992606074559185e-06,
|
| 1793 |
+
"loss": 29.315,
|
| 1794 |
+
"step": 2550
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.04750670962683572,
|
| 1798 |
+
"grad_norm": 31.65625,
|
| 1799 |
+
"learning_rate": 9.992577078773143e-06,
|
| 1800 |
+
"loss": 29.248,
|
| 1801 |
+
"step": 2560
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.04769228271131555,
|
| 1805 |
+
"grad_norm": 35.625,
|
| 1806 |
+
"learning_rate": 9.9925480829871e-06,
|
| 1807 |
+
"loss": 29.8038,
|
| 1808 |
+
"step": 2570
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.04787785579579538,
|
| 1812 |
+
"grad_norm": 33.625,
|
| 1813 |
+
"learning_rate": 9.992519087201058e-06,
|
| 1814 |
+
"loss": 29.1508,
|
| 1815 |
+
"step": 2580
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.048063428880275205,
|
| 1819 |
+
"grad_norm": 32.34375,
|
| 1820 |
+
"learning_rate": 9.992490091415015e-06,
|
| 1821 |
+
"loss": 29.4222,
|
| 1822 |
+
"step": 2590
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.04824900196475503,
|
| 1826 |
+
"grad_norm": 34.375,
|
| 1827 |
+
"learning_rate": 9.992461095628974e-06,
|
| 1828 |
+
"loss": 29.7663,
|
| 1829 |
+
"step": 2600
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.04843457504923486,
|
| 1833 |
+
"grad_norm": 34.5,
|
| 1834 |
+
"learning_rate": 9.99243209984293e-06,
|
| 1835 |
+
"loss": 29.1559,
|
| 1836 |
+
"step": 2610
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.04862014813371469,
|
| 1840 |
+
"grad_norm": 34.21875,
|
| 1841 |
+
"learning_rate": 9.992403104056887e-06,
|
| 1842 |
+
"loss": 29.0828,
|
| 1843 |
+
"step": 2620
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.048805721218194514,
|
| 1847 |
+
"grad_norm": 32.625,
|
| 1848 |
+
"learning_rate": 9.992374108270846e-06,
|
| 1849 |
+
"loss": 28.9075,
|
| 1850 |
+
"step": 2630
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.04899129430267434,
|
| 1854 |
+
"grad_norm": 33.46875,
|
| 1855 |
+
"learning_rate": 9.992345112484804e-06,
|
| 1856 |
+
"loss": 29.1731,
|
| 1857 |
+
"step": 2640
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.04917686738715417,
|
| 1861 |
+
"grad_norm": 33.5625,
|
| 1862 |
+
"learning_rate": 9.992316116698761e-06,
|
| 1863 |
+
"loss": 28.5537,
|
| 1864 |
+
"step": 2650
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.049362440471633996,
|
| 1868 |
+
"grad_norm": 32.59375,
|
| 1869 |
+
"learning_rate": 9.992287120912719e-06,
|
| 1870 |
+
"loss": 29.2992,
|
| 1871 |
+
"step": 2660
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.049548013556113824,
|
| 1875 |
+
"grad_norm": 33.6875,
|
| 1876 |
+
"learning_rate": 9.992258125126676e-06,
|
| 1877 |
+
"loss": 29.2843,
|
| 1878 |
+
"step": 2670
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.04973358664059365,
|
| 1882 |
+
"grad_norm": 35.5,
|
| 1883 |
+
"learning_rate": 9.992229129340633e-06,
|
| 1884 |
+
"loss": 29.039,
|
| 1885 |
+
"step": 2680
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.04991915972507348,
|
| 1889 |
+
"grad_norm": 31.34375,
|
| 1890 |
+
"learning_rate": 9.99220013355459e-06,
|
| 1891 |
+
"loss": 28.7605,
|
| 1892 |
+
"step": 2690
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.050104732809553305,
|
| 1896 |
+
"grad_norm": 32.65625,
|
| 1897 |
+
"learning_rate": 9.99217113776855e-06,
|
| 1898 |
+
"loss": 29.3325,
|
| 1899 |
+
"step": 2700
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.05029030589403313,
|
| 1903 |
+
"grad_norm": 33.71875,
|
| 1904 |
+
"learning_rate": 9.992142141982506e-06,
|
| 1905 |
+
"loss": 28.8713,
|
| 1906 |
+
"step": 2710
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.05047587897851295,
|
| 1910 |
+
"grad_norm": 31.328125,
|
| 1911 |
+
"learning_rate": 9.992113146196463e-06,
|
| 1912 |
+
"loss": 29.162,
|
| 1913 |
+
"step": 2720
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.05066145206299278,
|
| 1917 |
+
"grad_norm": 32.1875,
|
| 1918 |
+
"learning_rate": 9.992084150410422e-06,
|
| 1919 |
+
"loss": 28.794,
|
| 1920 |
+
"step": 2730
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.05084702514747261,
|
| 1924 |
+
"grad_norm": 33.78125,
|
| 1925 |
+
"learning_rate": 9.99205515462438e-06,
|
| 1926 |
+
"loss": 29.1353,
|
| 1927 |
+
"step": 2740
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.051032598231952435,
|
| 1931 |
+
"grad_norm": 33.125,
|
| 1932 |
+
"learning_rate": 9.992026158838337e-06,
|
| 1933 |
+
"loss": 29.5384,
|
| 1934 |
+
"step": 2750
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.05121817131643226,
|
| 1938 |
+
"grad_norm": 32.625,
|
| 1939 |
+
"learning_rate": 9.991997163052294e-06,
|
| 1940 |
+
"loss": 29.1446,
|
| 1941 |
+
"step": 2760
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.05140374440091209,
|
| 1945 |
+
"grad_norm": 34.03125,
|
| 1946 |
+
"learning_rate": 9.991968167266252e-06,
|
| 1947 |
+
"loss": 28.8186,
|
| 1948 |
+
"step": 2770
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.05158931748539192,
|
| 1952 |
+
"grad_norm": 31.96875,
|
| 1953 |
+
"learning_rate": 9.991939171480209e-06,
|
| 1954 |
+
"loss": 29.0019,
|
| 1955 |
+
"step": 2780
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.051774890569871744,
|
| 1959 |
+
"grad_norm": 31.84375,
|
| 1960 |
+
"learning_rate": 9.991910175694167e-06,
|
| 1961 |
+
"loss": 28.832,
|
| 1962 |
+
"step": 2790
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.05196046365435157,
|
| 1966 |
+
"grad_norm": 32.5625,
|
| 1967 |
+
"learning_rate": 9.991881179908126e-06,
|
| 1968 |
+
"loss": 28.9655,
|
| 1969 |
+
"step": 2800
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.0521460367388314,
|
| 1973 |
+
"grad_norm": 34.84375,
|
| 1974 |
+
"learning_rate": 9.991852184122083e-06,
|
| 1975 |
+
"loss": 28.7527,
|
| 1976 |
+
"step": 2810
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.052331609823311226,
|
| 1980 |
+
"grad_norm": 34.375,
|
| 1981 |
+
"learning_rate": 9.991823188336039e-06,
|
| 1982 |
+
"loss": 29.0628,
|
| 1983 |
+
"step": 2820
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.05251718290779105,
|
| 1987 |
+
"grad_norm": 31.453125,
|
| 1988 |
+
"learning_rate": 9.991794192549998e-06,
|
| 1989 |
+
"loss": 28.9482,
|
| 1990 |
+
"step": 2830
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.05270275599227088,
|
| 1994 |
+
"grad_norm": 35.15625,
|
| 1995 |
+
"learning_rate": 9.991765196763955e-06,
|
| 1996 |
+
"loss": 28.9719,
|
| 1997 |
+
"step": 2840
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.05288832907675071,
|
| 2001 |
+
"grad_norm": 31.28125,
|
| 2002 |
+
"learning_rate": 9.991736200977913e-06,
|
| 2003 |
+
"loss": 28.7786,
|
| 2004 |
+
"step": 2850
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.053073902161230535,
|
| 2008 |
+
"grad_norm": 33.15625,
|
| 2009 |
+
"learning_rate": 9.99170720519187e-06,
|
| 2010 |
+
"loss": 28.7459,
|
| 2011 |
+
"step": 2860
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.05325947524571036,
|
| 2015 |
+
"grad_norm": 33.5625,
|
| 2016 |
+
"learning_rate": 9.991678209405827e-06,
|
| 2017 |
+
"loss": 29.0512,
|
| 2018 |
+
"step": 2870
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.05344504833019019,
|
| 2022 |
+
"grad_norm": 34.75,
|
| 2023 |
+
"learning_rate": 9.991649213619785e-06,
|
| 2024 |
+
"loss": 28.7549,
|
| 2025 |
+
"step": 2880
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.05363062141467002,
|
| 2029 |
+
"grad_norm": 34.8125,
|
| 2030 |
+
"learning_rate": 9.991620217833742e-06,
|
| 2031 |
+
"loss": 29.0014,
|
| 2032 |
+
"step": 2890
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.053816194499149844,
|
| 2036 |
+
"grad_norm": 32.03125,
|
| 2037 |
+
"learning_rate": 9.991591222047701e-06,
|
| 2038 |
+
"loss": 28.7066,
|
| 2039 |
+
"step": 2900
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.05400176758362967,
|
| 2043 |
+
"grad_norm": 33.4375,
|
| 2044 |
+
"learning_rate": 9.991562226261659e-06,
|
| 2045 |
+
"loss": 28.8083,
|
| 2046 |
+
"step": 2910
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.0541873406681095,
|
| 2050 |
+
"grad_norm": 32.5,
|
| 2051 |
+
"learning_rate": 9.991533230475616e-06,
|
| 2052 |
+
"loss": 28.5716,
|
| 2053 |
+
"step": 2920
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.054372913752589326,
|
| 2057 |
+
"grad_norm": 32.53125,
|
| 2058 |
+
"learning_rate": 9.991504234689574e-06,
|
| 2059 |
+
"loss": 28.7969,
|
| 2060 |
+
"step": 2930
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.05455848683706915,
|
| 2064 |
+
"grad_norm": 33.09375,
|
| 2065 |
+
"learning_rate": 9.991475238903531e-06,
|
| 2066 |
+
"loss": 28.535,
|
| 2067 |
+
"step": 2940
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.05474405992154898,
|
| 2071 |
+
"grad_norm": 32.5625,
|
| 2072 |
+
"learning_rate": 9.991446243117488e-06,
|
| 2073 |
+
"loss": 28.7149,
|
| 2074 |
+
"step": 2950
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.05492963300602881,
|
| 2078 |
+
"grad_norm": 33.75,
|
| 2079 |
+
"learning_rate": 9.991417247331446e-06,
|
| 2080 |
+
"loss": 28.7103,
|
| 2081 |
+
"step": 2960
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.055115206090508635,
|
| 2085 |
+
"grad_norm": 32.53125,
|
| 2086 |
+
"learning_rate": 9.991388251545405e-06,
|
| 2087 |
+
"loss": 28.6576,
|
| 2088 |
+
"step": 2970
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.05530077917498846,
|
| 2092 |
+
"grad_norm": 35.375,
|
| 2093 |
+
"learning_rate": 9.99135925575936e-06,
|
| 2094 |
+
"loss": 28.6555,
|
| 2095 |
+
"step": 2980
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.05548635225946829,
|
| 2099 |
+
"grad_norm": 33.78125,
|
| 2100 |
+
"learning_rate": 9.991330259973318e-06,
|
| 2101 |
+
"loss": 28.4429,
|
| 2102 |
+
"step": 2990
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.05567192534394812,
|
| 2106 |
+
"grad_norm": 33.3125,
|
| 2107 |
+
"learning_rate": 9.991301264187277e-06,
|
| 2108 |
+
"loss": 28.3745,
|
| 2109 |
+
"step": 3000
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.05585749842842794,
|
| 2113 |
+
"grad_norm": 33.5,
|
| 2114 |
+
"learning_rate": 9.991272268401235e-06,
|
| 2115 |
+
"loss": 28.3746,
|
| 2116 |
+
"step": 3010
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.056043071512907765,
|
| 2120 |
+
"grad_norm": 31.96875,
|
| 2121 |
+
"learning_rate": 9.991243272615192e-06,
|
| 2122 |
+
"loss": 28.5534,
|
| 2123 |
+
"step": 3020
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.05622864459738759,
|
| 2127 |
+
"grad_norm": 34.25,
|
| 2128 |
+
"learning_rate": 9.99121427682915e-06,
|
| 2129 |
+
"loss": 28.6145,
|
| 2130 |
+
"step": 3030
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.05641421768186742,
|
| 2134 |
+
"grad_norm": 33.46875,
|
| 2135 |
+
"learning_rate": 9.991185281043107e-06,
|
| 2136 |
+
"loss": 28.6077,
|
| 2137 |
+
"step": 3040
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.05659979076634725,
|
| 2141 |
+
"grad_norm": 33.375,
|
| 2142 |
+
"learning_rate": 9.991156285257064e-06,
|
| 2143 |
+
"loss": 28.967,
|
| 2144 |
+
"step": 3050
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.056785363850827074,
|
| 2148 |
+
"grad_norm": 33.1875,
|
| 2149 |
+
"learning_rate": 9.991127289471022e-06,
|
| 2150 |
+
"loss": 28.4267,
|
| 2151 |
+
"step": 3060
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.0569709369353069,
|
| 2155 |
+
"grad_norm": 33.53125,
|
| 2156 |
+
"learning_rate": 9.991098293684979e-06,
|
| 2157 |
+
"loss": 28.6182,
|
| 2158 |
+
"step": 3070
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.05715651001978673,
|
| 2162 |
+
"grad_norm": 33.03125,
|
| 2163 |
+
"learning_rate": 9.991069297898938e-06,
|
| 2164 |
+
"loss": 28.5312,
|
| 2165 |
+
"step": 3080
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.057342083104266556,
|
| 2169 |
+
"grad_norm": 33.28125,
|
| 2170 |
+
"learning_rate": 9.991040302112894e-06,
|
| 2171 |
+
"loss": 28.7762,
|
| 2172 |
+
"step": 3090
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.05752765618874638,
|
| 2176 |
+
"grad_norm": 33.84375,
|
| 2177 |
+
"learning_rate": 9.991011306326853e-06,
|
| 2178 |
+
"loss": 28.4776,
|
| 2179 |
+
"step": 3100
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.05771322927322621,
|
| 2183 |
+
"grad_norm": 31.390625,
|
| 2184 |
+
"learning_rate": 9.99098231054081e-06,
|
| 2185 |
+
"loss": 28.2534,
|
| 2186 |
+
"step": 3110
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.05789880235770604,
|
| 2190 |
+
"grad_norm": 33.0,
|
| 2191 |
+
"learning_rate": 9.990953314754768e-06,
|
| 2192 |
+
"loss": 28.5445,
|
| 2193 |
+
"step": 3120
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.058084375442185865,
|
| 2197 |
+
"grad_norm": 32.8125,
|
| 2198 |
+
"learning_rate": 9.990924318968725e-06,
|
| 2199 |
+
"loss": 28.4319,
|
| 2200 |
+
"step": 3130
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.05826994852666569,
|
| 2204 |
+
"grad_norm": 32.9375,
|
| 2205 |
+
"learning_rate": 9.990895323182683e-06,
|
| 2206 |
+
"loss": 28.4076,
|
| 2207 |
+
"step": 3140
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.05845552161114552,
|
| 2211 |
+
"grad_norm": 34.09375,
|
| 2212 |
+
"learning_rate": 9.99086632739664e-06,
|
| 2213 |
+
"loss": 28.4729,
|
| 2214 |
+
"step": 3150
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.05864109469562535,
|
| 2218 |
+
"grad_norm": 34.09375,
|
| 2219 |
+
"learning_rate": 9.990837331610597e-06,
|
| 2220 |
+
"loss": 28.3709,
|
| 2221 |
+
"step": 3160
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.058826667780105174,
|
| 2225 |
+
"grad_norm": 33.0,
|
| 2226 |
+
"learning_rate": 9.990808335824555e-06,
|
| 2227 |
+
"loss": 28.149,
|
| 2228 |
+
"step": 3170
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.059012240864585,
|
| 2232 |
+
"grad_norm": 33.21875,
|
| 2233 |
+
"learning_rate": 9.990779340038514e-06,
|
| 2234 |
+
"loss": 28.3831,
|
| 2235 |
+
"step": 3180
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.05919781394906483,
|
| 2239 |
+
"grad_norm": 33.5,
|
| 2240 |
+
"learning_rate": 9.990750344252471e-06,
|
| 2241 |
+
"loss": 28.5165,
|
| 2242 |
+
"step": 3190
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.059383387033544656,
|
| 2246 |
+
"grad_norm": 32.71875,
|
| 2247 |
+
"learning_rate": 9.990721348466427e-06,
|
| 2248 |
+
"loss": 28.5913,
|
| 2249 |
+
"step": 3200
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.05956896011802448,
|
| 2253 |
+
"grad_norm": 32.40625,
|
| 2254 |
+
"learning_rate": 9.990692352680386e-06,
|
| 2255 |
+
"loss": 27.9974,
|
| 2256 |
+
"step": 3210
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.05975453320250431,
|
| 2260 |
+
"grad_norm": 35.0,
|
| 2261 |
+
"learning_rate": 9.990663356894344e-06,
|
| 2262 |
+
"loss": 28.6096,
|
| 2263 |
+
"step": 3220
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.05994010628698414,
|
| 2267 |
+
"grad_norm": 33.6875,
|
| 2268 |
+
"learning_rate": 9.990634361108301e-06,
|
| 2269 |
+
"loss": 28.4732,
|
| 2270 |
+
"step": 3230
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.060125679371463965,
|
| 2274 |
+
"grad_norm": 34.3125,
|
| 2275 |
+
"learning_rate": 9.990605365322258e-06,
|
| 2276 |
+
"loss": 28.7017,
|
| 2277 |
+
"step": 3240
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.06031125245594379,
|
| 2281 |
+
"grad_norm": 34.09375,
|
| 2282 |
+
"learning_rate": 9.990576369536216e-06,
|
| 2283 |
+
"loss": 28.2992,
|
| 2284 |
+
"step": 3250
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.06049682554042362,
|
| 2288 |
+
"grad_norm": 36.84375,
|
| 2289 |
+
"learning_rate": 9.990547373750173e-06,
|
| 2290 |
+
"loss": 28.4159,
|
| 2291 |
+
"step": 3260
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.06068239862490345,
|
| 2295 |
+
"grad_norm": 33.4375,
|
| 2296 |
+
"learning_rate": 9.99051837796413e-06,
|
| 2297 |
+
"loss": 28.3044,
|
| 2298 |
+
"step": 3270
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.060867971709383274,
|
| 2302 |
+
"grad_norm": 32.96875,
|
| 2303 |
+
"learning_rate": 9.99048938217809e-06,
|
| 2304 |
+
"loss": 28.8831,
|
| 2305 |
+
"step": 3280
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.0610535447938631,
|
| 2309 |
+
"grad_norm": 34.0,
|
| 2310 |
+
"learning_rate": 9.990460386392047e-06,
|
| 2311 |
+
"loss": 28.2866,
|
| 2312 |
+
"step": 3290
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.06123911787834292,
|
| 2316 |
+
"grad_norm": 32.59375,
|
| 2317 |
+
"learning_rate": 9.990431390606003e-06,
|
| 2318 |
+
"loss": 28.5798,
|
| 2319 |
+
"step": 3300
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.06142469096282275,
|
| 2323 |
+
"grad_norm": 35.59375,
|
| 2324 |
+
"learning_rate": 9.990402394819962e-06,
|
| 2325 |
+
"loss": 28.379,
|
| 2326 |
+
"step": 3310
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.061610264047302576,
|
| 2330 |
+
"grad_norm": 32.4375,
|
| 2331 |
+
"learning_rate": 9.99037339903392e-06,
|
| 2332 |
+
"loss": 27.8911,
|
| 2333 |
+
"step": 3320
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.061795837131782404,
|
| 2337 |
+
"grad_norm": 31.421875,
|
| 2338 |
+
"learning_rate": 9.990344403247877e-06,
|
| 2339 |
+
"loss": 27.7105,
|
| 2340 |
+
"step": 3330
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.06198141021626223,
|
| 2344 |
+
"grad_norm": 34.90625,
|
| 2345 |
+
"learning_rate": 9.990315407461834e-06,
|
| 2346 |
+
"loss": 27.8929,
|
| 2347 |
+
"step": 3340
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.06216698330074206,
|
| 2351 |
+
"grad_norm": 31.84375,
|
| 2352 |
+
"learning_rate": 9.990286411675793e-06,
|
| 2353 |
+
"loss": 28.0276,
|
| 2354 |
+
"step": 3350
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.062352556385221886,
|
| 2358 |
+
"grad_norm": 33.21875,
|
| 2359 |
+
"learning_rate": 9.990257415889749e-06,
|
| 2360 |
+
"loss": 28.489,
|
| 2361 |
+
"step": 3360
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.06253812946970172,
|
| 2365 |
+
"grad_norm": 32.96875,
|
| 2366 |
+
"learning_rate": 9.990228420103706e-06,
|
| 2367 |
+
"loss": 28.0725,
|
| 2368 |
+
"step": 3370
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.06272370255418154,
|
| 2372 |
+
"grad_norm": 32.90625,
|
| 2373 |
+
"learning_rate": 9.990199424317665e-06,
|
| 2374 |
+
"loss": 28.3279,
|
| 2375 |
+
"step": 3380
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.06290927563866137,
|
| 2379 |
+
"grad_norm": 34.21875,
|
| 2380 |
+
"learning_rate": 9.990170428531623e-06,
|
| 2381 |
+
"loss": 28.3579,
|
| 2382 |
+
"step": 3390
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.0630948487231412,
|
| 2386 |
+
"grad_norm": 35.1875,
|
| 2387 |
+
"learning_rate": 9.99014143274558e-06,
|
| 2388 |
+
"loss": 28.056,
|
| 2389 |
+
"step": 3400
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.06328042180762103,
|
| 2393 |
+
"grad_norm": 34.25,
|
| 2394 |
+
"learning_rate": 9.990112436959538e-06,
|
| 2395 |
+
"loss": 28.0313,
|
| 2396 |
+
"step": 3410
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.06346599489210085,
|
| 2400 |
+
"grad_norm": 33.28125,
|
| 2401 |
+
"learning_rate": 9.990083441173495e-06,
|
| 2402 |
+
"loss": 28.1424,
|
| 2403 |
+
"step": 3420
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.06365156797658068,
|
| 2407 |
+
"grad_norm": 33.25,
|
| 2408 |
+
"learning_rate": 9.990054445387452e-06,
|
| 2409 |
+
"loss": 28.0296,
|
| 2410 |
+
"step": 3430
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.0638371410610605,
|
| 2414 |
+
"grad_norm": 32.75,
|
| 2415 |
+
"learning_rate": 9.99002544960141e-06,
|
| 2416 |
+
"loss": 27.9722,
|
| 2417 |
+
"step": 3440
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.06402271414554032,
|
| 2421 |
+
"grad_norm": 33.5625,
|
| 2422 |
+
"learning_rate": 9.989996453815369e-06,
|
| 2423 |
+
"loss": 27.9661,
|
| 2424 |
+
"step": 3450
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.06420828723002016,
|
| 2428 |
+
"grad_norm": 34.0625,
|
| 2429 |
+
"learning_rate": 9.989967458029325e-06,
|
| 2430 |
+
"loss": 28.3075,
|
| 2431 |
+
"step": 3460
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.06439386031449998,
|
| 2435 |
+
"grad_norm": 32.6875,
|
| 2436 |
+
"learning_rate": 9.989938462243282e-06,
|
| 2437 |
+
"loss": 27.7058,
|
| 2438 |
+
"step": 3470
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.06457943339897981,
|
| 2442 |
+
"grad_norm": 33.90625,
|
| 2443 |
+
"learning_rate": 9.989909466457241e-06,
|
| 2444 |
+
"loss": 28.327,
|
| 2445 |
+
"step": 3480
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.06476500648345963,
|
| 2449 |
+
"grad_norm": 35.5625,
|
| 2450 |
+
"learning_rate": 9.989880470671199e-06,
|
| 2451 |
+
"loss": 28.1936,
|
| 2452 |
+
"step": 3490
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.06495057956793947,
|
| 2456 |
+
"grad_norm": 34.375,
|
| 2457 |
+
"learning_rate": 9.989851474885156e-06,
|
| 2458 |
+
"loss": 28.0495,
|
| 2459 |
+
"step": 3500
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.06513615265241929,
|
| 2463 |
+
"grad_norm": 33.8125,
|
| 2464 |
+
"learning_rate": 9.989822479099113e-06,
|
| 2465 |
+
"loss": 27.782,
|
| 2466 |
+
"step": 3510
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.06532172573689912,
|
| 2470 |
+
"grad_norm": 33.53125,
|
| 2471 |
+
"learning_rate": 9.98979348331307e-06,
|
| 2472 |
+
"loss": 27.8067,
|
| 2473 |
+
"step": 3520
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.06550729882137894,
|
| 2477 |
+
"grad_norm": 32.71875,
|
| 2478 |
+
"learning_rate": 9.989764487527028e-06,
|
| 2479 |
+
"loss": 27.7428,
|
| 2480 |
+
"step": 3530
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.06569287190585878,
|
| 2484 |
+
"grad_norm": 34.125,
|
| 2485 |
+
"learning_rate": 9.989735491740986e-06,
|
| 2486 |
+
"loss": 27.872,
|
| 2487 |
+
"step": 3540
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.0658784449903386,
|
| 2491 |
+
"grad_norm": 34.875,
|
| 2492 |
+
"learning_rate": 9.989706495954945e-06,
|
| 2493 |
+
"loss": 27.917,
|
| 2494 |
+
"step": 3550
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.06606401807481843,
|
| 2498 |
+
"grad_norm": 33.75,
|
| 2499 |
+
"learning_rate": 9.989677500168902e-06,
|
| 2500 |
+
"loss": 27.772,
|
| 2501 |
+
"step": 3560
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.06624959115929825,
|
| 2505 |
+
"grad_norm": 34.125,
|
| 2506 |
+
"learning_rate": 9.989648504382858e-06,
|
| 2507 |
+
"loss": 27.3048,
|
| 2508 |
+
"step": 3570
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.06643516424377809,
|
| 2512 |
+
"grad_norm": 35.59375,
|
| 2513 |
+
"learning_rate": 9.989619508596817e-06,
|
| 2514 |
+
"loss": 27.7842,
|
| 2515 |
+
"step": 3580
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.0666207373282579,
|
| 2519 |
+
"grad_norm": 33.125,
|
| 2520 |
+
"learning_rate": 9.989590512810774e-06,
|
| 2521 |
+
"loss": 27.7142,
|
| 2522 |
+
"step": 3590
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.06680631041273774,
|
| 2526 |
+
"grad_norm": 34.875,
|
| 2527 |
+
"learning_rate": 9.989561517024732e-06,
|
| 2528 |
+
"loss": 27.9764,
|
| 2529 |
+
"step": 3600
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.06699188349721756,
|
| 2533 |
+
"grad_norm": 33.03125,
|
| 2534 |
+
"learning_rate": 9.98953252123869e-06,
|
| 2535 |
+
"loss": 27.9061,
|
| 2536 |
+
"step": 3610
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.0671774565816974,
|
| 2540 |
+
"grad_norm": 34.09375,
|
| 2541 |
+
"learning_rate": 9.989503525452647e-06,
|
| 2542 |
+
"loss": 28.1396,
|
| 2543 |
+
"step": 3620
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.06736302966617722,
|
| 2547 |
+
"grad_norm": 34.375,
|
| 2548 |
+
"learning_rate": 9.989474529666604e-06,
|
| 2549 |
+
"loss": 27.8155,
|
| 2550 |
+
"step": 3630
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.06754860275065705,
|
| 2554 |
+
"grad_norm": 33.59375,
|
| 2555 |
+
"learning_rate": 9.989445533880561e-06,
|
| 2556 |
+
"loss": 28.192,
|
| 2557 |
+
"step": 3640
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.06773417583513687,
|
| 2561 |
+
"grad_norm": 33.40625,
|
| 2562 |
+
"learning_rate": 9.989416538094519e-06,
|
| 2563 |
+
"loss": 27.6073,
|
| 2564 |
+
"step": 3650
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.0679197489196167,
|
| 2568 |
+
"grad_norm": 33.65625,
|
| 2569 |
+
"learning_rate": 9.989387542308478e-06,
|
| 2570 |
+
"loss": 27.8798,
|
| 2571 |
+
"step": 3660
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.06810532200409652,
|
| 2575 |
+
"grad_norm": 34.0,
|
| 2576 |
+
"learning_rate": 9.989358546522435e-06,
|
| 2577 |
+
"loss": 27.4086,
|
| 2578 |
+
"step": 3670
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.06829089508857636,
|
| 2582 |
+
"grad_norm": 34.3125,
|
| 2583 |
+
"learning_rate": 9.989329550736393e-06,
|
| 2584 |
+
"loss": 28.0494,
|
| 2585 |
+
"step": 3680
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.06847646817305618,
|
| 2589 |
+
"grad_norm": 33.71875,
|
| 2590 |
+
"learning_rate": 9.98930055495035e-06,
|
| 2591 |
+
"loss": 27.7582,
|
| 2592 |
+
"step": 3690
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.06866204125753601,
|
| 2596 |
+
"grad_norm": 32.75,
|
| 2597 |
+
"learning_rate": 9.989271559164308e-06,
|
| 2598 |
+
"loss": 27.7977,
|
| 2599 |
+
"step": 3700
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.06884761434201583,
|
| 2603 |
+
"grad_norm": 33.34375,
|
| 2604 |
+
"learning_rate": 9.989242563378265e-06,
|
| 2605 |
+
"loss": 28.2126,
|
| 2606 |
+
"step": 3710
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.06903318742649567,
|
| 2610 |
+
"grad_norm": 31.671875,
|
| 2611 |
+
"learning_rate": 9.989213567592222e-06,
|
| 2612 |
+
"loss": 27.4855,
|
| 2613 |
+
"step": 3720
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.06921876051097549,
|
| 2617 |
+
"grad_norm": 32.65625,
|
| 2618 |
+
"learning_rate": 9.98918457180618e-06,
|
| 2619 |
+
"loss": 27.8408,
|
| 2620 |
+
"step": 3730
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.06940433359545531,
|
| 2624 |
+
"grad_norm": 31.484375,
|
| 2625 |
+
"learning_rate": 9.989155576020137e-06,
|
| 2626 |
+
"loss": 27.7657,
|
| 2627 |
+
"step": 3740
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.06958990667993514,
|
| 2631 |
+
"grad_norm": 33.1875,
|
| 2632 |
+
"learning_rate": 9.989126580234095e-06,
|
| 2633 |
+
"loss": 27.8898,
|
| 2634 |
+
"step": 3750
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.06977547976441496,
|
| 2638 |
+
"grad_norm": 32.90625,
|
| 2639 |
+
"learning_rate": 9.989097584448054e-06,
|
| 2640 |
+
"loss": 27.4146,
|
| 2641 |
+
"step": 3760
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.0699610528488948,
|
| 2645 |
+
"grad_norm": 31.53125,
|
| 2646 |
+
"learning_rate": 9.989068588662011e-06,
|
| 2647 |
+
"loss": 27.6763,
|
| 2648 |
+
"step": 3770
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.07014662593337462,
|
| 2652 |
+
"grad_norm": 33.125,
|
| 2653 |
+
"learning_rate": 9.989039592875968e-06,
|
| 2654 |
+
"loss": 27.6253,
|
| 2655 |
+
"step": 3780
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.07033219901785445,
|
| 2659 |
+
"grad_norm": 35.0,
|
| 2660 |
+
"learning_rate": 9.989010597089926e-06,
|
| 2661 |
+
"loss": 27.7828,
|
| 2662 |
+
"step": 3790
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 0.07051777210233427,
|
| 2666 |
+
"grad_norm": 35.84375,
|
| 2667 |
+
"learning_rate": 9.988981601303883e-06,
|
| 2668 |
+
"loss": 27.6479,
|
| 2669 |
+
"step": 3800
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 0.0707033451868141,
|
| 2673 |
+
"grad_norm": 33.03125,
|
| 2674 |
+
"learning_rate": 9.98895260551784e-06,
|
| 2675 |
+
"loss": 27.5623,
|
| 2676 |
+
"step": 3810
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.07088891827129393,
|
| 2680 |
+
"grad_norm": 32.84375,
|
| 2681 |
+
"learning_rate": 9.988923609731798e-06,
|
| 2682 |
+
"loss": 27.1696,
|
| 2683 |
+
"step": 3820
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.07107449135577376,
|
| 2687 |
+
"grad_norm": 33.84375,
|
| 2688 |
+
"learning_rate": 9.988894613945757e-06,
|
| 2689 |
+
"loss": 27.4184,
|
| 2690 |
+
"step": 3830
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.07126006444025358,
|
| 2694 |
+
"grad_norm": 34.5625,
|
| 2695 |
+
"learning_rate": 9.988865618159713e-06,
|
| 2696 |
+
"loss": 27.6026,
|
| 2697 |
+
"step": 3840
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 0.07144563752473342,
|
| 2701 |
+
"grad_norm": 35.6875,
|
| 2702 |
+
"learning_rate": 9.98883662237367e-06,
|
| 2703 |
+
"loss": 27.5197,
|
| 2704 |
+
"step": 3850
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 0.07163121060921324,
|
| 2708 |
+
"grad_norm": 33.15625,
|
| 2709 |
+
"learning_rate": 9.98880762658763e-06,
|
| 2710 |
+
"loss": 27.482,
|
| 2711 |
+
"step": 3860
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 0.07181678369369307,
|
| 2715 |
+
"grad_norm": 33.90625,
|
| 2716 |
+
"learning_rate": 9.988778630801587e-06,
|
| 2717 |
+
"loss": 27.7027,
|
| 2718 |
+
"step": 3870
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 0.07200235677817289,
|
| 2722 |
+
"grad_norm": 32.375,
|
| 2723 |
+
"learning_rate": 9.988749635015544e-06,
|
| 2724 |
+
"loss": 27.3702,
|
| 2725 |
+
"step": 3880
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 0.07218792986265272,
|
| 2729 |
+
"grad_norm": 34.03125,
|
| 2730 |
+
"learning_rate": 9.988720639229502e-06,
|
| 2731 |
+
"loss": 26.946,
|
| 2732 |
+
"step": 3890
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 0.07237350294713255,
|
| 2736 |
+
"grad_norm": 36.34375,
|
| 2737 |
+
"learning_rate": 9.988691643443459e-06,
|
| 2738 |
+
"loss": 27.8434,
|
| 2739 |
+
"step": 3900
|
| 2740 |
+
},
|
| 2741 |
+
{
|
| 2742 |
+
"epoch": 0.07255907603161238,
|
| 2743 |
+
"grad_norm": 33.8125,
|
| 2744 |
+
"learning_rate": 9.988662647657416e-06,
|
| 2745 |
+
"loss": 27.6265,
|
| 2746 |
+
"step": 3910
|
| 2747 |
+
},
|
| 2748 |
+
{
|
| 2749 |
+
"epoch": 0.0727446491160922,
|
| 2750 |
+
"grad_norm": 34.59375,
|
| 2751 |
+
"learning_rate": 9.988633651871374e-06,
|
| 2752 |
+
"loss": 27.367,
|
| 2753 |
+
"step": 3920
|
| 2754 |
+
},
|
| 2755 |
+
{
|
| 2756 |
+
"epoch": 0.07293022220057203,
|
| 2757 |
+
"grad_norm": 35.15625,
|
| 2758 |
+
"learning_rate": 9.988604656085333e-06,
|
| 2759 |
+
"loss": 27.686,
|
| 2760 |
+
"step": 3930
|
| 2761 |
+
},
|
| 2762 |
+
{
|
| 2763 |
+
"epoch": 0.07311579528505185,
|
| 2764 |
+
"grad_norm": 32.625,
|
| 2765 |
+
"learning_rate": 9.98857566029929e-06,
|
| 2766 |
+
"loss": 27.4221,
|
| 2767 |
+
"step": 3940
|
| 2768 |
+
},
|
| 2769 |
+
{
|
| 2770 |
+
"epoch": 0.07330136836953169,
|
| 2771 |
+
"grad_norm": 32.75,
|
| 2772 |
+
"learning_rate": 9.988546664513246e-06,
|
| 2773 |
+
"loss": 27.7722,
|
| 2774 |
+
"step": 3950
|
| 2775 |
+
},
|
| 2776 |
+
{
|
| 2777 |
+
"epoch": 0.07348694145401151,
|
| 2778 |
+
"grad_norm": 34.78125,
|
| 2779 |
+
"learning_rate": 9.988517668727205e-06,
|
| 2780 |
+
"loss": 27.3738,
|
| 2781 |
+
"step": 3960
|
| 2782 |
+
},
|
| 2783 |
+
{
|
| 2784 |
+
"epoch": 0.07367251453849134,
|
| 2785 |
+
"grad_norm": 33.9375,
|
| 2786 |
+
"learning_rate": 9.988488672941163e-06,
|
| 2787 |
+
"loss": 27.7157,
|
| 2788 |
+
"step": 3970
|
| 2789 |
+
},
|
| 2790 |
+
{
|
| 2791 |
+
"epoch": 0.07385808762297116,
|
| 2792 |
+
"grad_norm": 34.59375,
|
| 2793 |
+
"learning_rate": 9.98845967715512e-06,
|
| 2794 |
+
"loss": 27.328,
|
| 2795 |
+
"step": 3980
|
| 2796 |
+
},
|
| 2797 |
+
{
|
| 2798 |
+
"epoch": 0.074043660707451,
|
| 2799 |
+
"grad_norm": 34.34375,
|
| 2800 |
+
"learning_rate": 9.988430681369077e-06,
|
| 2801 |
+
"loss": 27.4072,
|
| 2802 |
+
"step": 3990
|
| 2803 |
+
},
|
| 2804 |
+
{
|
| 2805 |
+
"epoch": 0.07422923379193082,
|
| 2806 |
+
"grad_norm": 32.96875,
|
| 2807 |
+
"learning_rate": 9.988401685583035e-06,
|
| 2808 |
+
"loss": 27.6186,
|
| 2809 |
+
"step": 4000
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 0.07441480687641064,
|
| 2813 |
+
"grad_norm": 33.84375,
|
| 2814 |
+
"learning_rate": 9.988372689796992e-06,
|
| 2815 |
+
"loss": 27.6838,
|
| 2816 |
+
"step": 4010
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 0.07460037996089047,
|
| 2820 |
+
"grad_norm": 33.0625,
|
| 2821 |
+
"learning_rate": 9.98834369401095e-06,
|
| 2822 |
+
"loss": 27.4932,
|
| 2823 |
+
"step": 4020
|
| 2824 |
+
},
|
| 2825 |
+
{
|
| 2826 |
+
"epoch": 0.0747859530453703,
|
| 2827 |
+
"grad_norm": 34.375,
|
| 2828 |
+
"learning_rate": 9.988314698224909e-06,
|
| 2829 |
+
"loss": 27.4686,
|
| 2830 |
+
"step": 4030
|
| 2831 |
+
},
|
| 2832 |
+
{
|
| 2833 |
+
"epoch": 0.07497152612985013,
|
| 2834 |
+
"grad_norm": 34.8125,
|
| 2835 |
+
"learning_rate": 9.988285702438866e-06,
|
| 2836 |
+
"loss": 27.094,
|
| 2837 |
+
"step": 4040
|
| 2838 |
+
},
|
| 2839 |
+
{
|
| 2840 |
+
"epoch": 0.07515709921432995,
|
| 2841 |
+
"grad_norm": 32.25,
|
| 2842 |
+
"learning_rate": 9.988256706652822e-06,
|
| 2843 |
+
"loss": 27.2472,
|
| 2844 |
+
"step": 4050
|
| 2845 |
+
},
|
| 2846 |
+
{
|
| 2847 |
+
"epoch": 0.07534267229880978,
|
| 2848 |
+
"grad_norm": 33.0,
|
| 2849 |
+
"learning_rate": 9.988227710866781e-06,
|
| 2850 |
+
"loss": 27.2044,
|
| 2851 |
+
"step": 4060
|
| 2852 |
+
},
|
| 2853 |
+
{
|
| 2854 |
+
"epoch": 0.0755282453832896,
|
| 2855 |
+
"grad_norm": 32.84375,
|
| 2856 |
+
"learning_rate": 9.988198715080738e-06,
|
| 2857 |
+
"loss": 27.0425,
|
| 2858 |
+
"step": 4070
|
| 2859 |
+
},
|
| 2860 |
+
{
|
| 2861 |
+
"epoch": 0.07571381846776944,
|
| 2862 |
+
"grad_norm": 36.15625,
|
| 2863 |
+
"learning_rate": 9.988169719294696e-06,
|
| 2864 |
+
"loss": 27.5593,
|
| 2865 |
+
"step": 4080
|
| 2866 |
+
},
|
| 2867 |
+
{
|
| 2868 |
+
"epoch": 0.07589939155224926,
|
| 2869 |
+
"grad_norm": 32.96875,
|
| 2870 |
+
"learning_rate": 9.988140723508653e-06,
|
| 2871 |
+
"loss": 26.8557,
|
| 2872 |
+
"step": 4090
|
| 2873 |
+
},
|
| 2874 |
+
{
|
| 2875 |
+
"epoch": 0.07608496463672909,
|
| 2876 |
+
"grad_norm": 33.25,
|
| 2877 |
+
"learning_rate": 9.98811172772261e-06,
|
| 2878 |
+
"loss": 27.2413,
|
| 2879 |
+
"step": 4100
|
| 2880 |
+
},
|
| 2881 |
+
{
|
| 2882 |
+
"epoch": 0.07627053772120891,
|
| 2883 |
+
"grad_norm": 33.90625,
|
| 2884 |
+
"learning_rate": 9.988082731936568e-06,
|
| 2885 |
+
"loss": 27.032,
|
| 2886 |
+
"step": 4110
|
| 2887 |
+
},
|
| 2888 |
+
{
|
| 2889 |
+
"epoch": 0.07645611080568875,
|
| 2890 |
+
"grad_norm": 34.15625,
|
| 2891 |
+
"learning_rate": 9.988053736150525e-06,
|
| 2892 |
+
"loss": 27.1695,
|
| 2893 |
+
"step": 4120
|
| 2894 |
+
},
|
| 2895 |
+
{
|
| 2896 |
+
"epoch": 0.07664168389016857,
|
| 2897 |
+
"grad_norm": 33.59375,
|
| 2898 |
+
"learning_rate": 9.988024740364485e-06,
|
| 2899 |
+
"loss": 27.4066,
|
| 2900 |
+
"step": 4130
|
| 2901 |
+
},
|
| 2902 |
+
{
|
| 2903 |
+
"epoch": 0.0768272569746484,
|
| 2904 |
+
"grad_norm": 36.03125,
|
| 2905 |
+
"learning_rate": 9.987995744578442e-06,
|
| 2906 |
+
"loss": 27.7367,
|
| 2907 |
+
"step": 4140
|
| 2908 |
+
},
|
| 2909 |
+
{
|
| 2910 |
+
"epoch": 0.07701283005912822,
|
| 2911 |
+
"grad_norm": 35.375,
|
| 2912 |
+
"learning_rate": 9.9879667487924e-06,
|
| 2913 |
+
"loss": 27.0799,
|
| 2914 |
+
"step": 4150
|
| 2915 |
+
},
|
| 2916 |
+
{
|
| 2917 |
+
"epoch": 0.07719840314360805,
|
| 2918 |
+
"grad_norm": 34.21875,
|
| 2919 |
+
"learning_rate": 9.987937753006357e-06,
|
| 2920 |
+
"loss": 27.4982,
|
| 2921 |
+
"step": 4160
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 0.07738397622808788,
|
| 2925 |
+
"grad_norm": 33.15625,
|
| 2926 |
+
"learning_rate": 9.987908757220314e-06,
|
| 2927 |
+
"loss": 26.987,
|
| 2928 |
+
"step": 4170
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 0.07756954931256771,
|
| 2932 |
+
"grad_norm": 32.53125,
|
| 2933 |
+
"learning_rate": 9.987879761434272e-06,
|
| 2934 |
+
"loss": 27.2336,
|
| 2935 |
+
"step": 4180
|
| 2936 |
+
},
|
| 2937 |
+
{
|
| 2938 |
+
"epoch": 0.07775512239704753,
|
| 2939 |
+
"grad_norm": 34.96875,
|
| 2940 |
+
"learning_rate": 9.987850765648229e-06,
|
| 2941 |
+
"loss": 27.14,
|
| 2942 |
+
"step": 4190
|
| 2943 |
+
},
|
| 2944 |
+
{
|
| 2945 |
+
"epoch": 0.07794069548152736,
|
| 2946 |
+
"grad_norm": 34.8125,
|
| 2947 |
+
"learning_rate": 9.987821769862186e-06,
|
| 2948 |
+
"loss": 27.8387,
|
| 2949 |
+
"step": 4200
|
| 2950 |
+
},
|
| 2951 |
+
{
|
| 2952 |
+
"epoch": 0.07812626856600718,
|
| 2953 |
+
"grad_norm": 34.09375,
|
| 2954 |
+
"learning_rate": 9.987792774076144e-06,
|
| 2955 |
+
"loss": 27.4868,
|
| 2956 |
+
"step": 4210
|
| 2957 |
+
},
|
| 2958 |
+
{
|
| 2959 |
+
"epoch": 0.07831184165048702,
|
| 2960 |
+
"grad_norm": 33.875,
|
| 2961 |
+
"learning_rate": 9.987763778290101e-06,
|
| 2962 |
+
"loss": 27.4351,
|
| 2963 |
+
"step": 4220
|
| 2964 |
+
},
|
| 2965 |
+
{
|
| 2966 |
+
"epoch": 0.07849741473496684,
|
| 2967 |
+
"grad_norm": 35.21875,
|
| 2968 |
+
"learning_rate": 9.987734782504059e-06,
|
| 2969 |
+
"loss": 27.3241,
|
| 2970 |
+
"step": 4230
|
| 2971 |
+
},
|
| 2972 |
+
{
|
| 2973 |
+
"epoch": 0.07868298781944667,
|
| 2974 |
+
"grad_norm": 33.0,
|
| 2975 |
+
"learning_rate": 9.987705786718018e-06,
|
| 2976 |
+
"loss": 27.6498,
|
| 2977 |
+
"step": 4240
|
| 2978 |
+
},
|
| 2979 |
+
{
|
| 2980 |
+
"epoch": 0.0788685609039265,
|
| 2981 |
+
"grad_norm": 33.09375,
|
| 2982 |
+
"learning_rate": 9.987676790931975e-06,
|
| 2983 |
+
"loss": 27.1418,
|
| 2984 |
+
"step": 4250
|
| 2985 |
+
},
|
| 2986 |
+
{
|
| 2987 |
+
"epoch": 0.07905413398840633,
|
| 2988 |
+
"grad_norm": 32.8125,
|
| 2989 |
+
"learning_rate": 9.987647795145932e-06,
|
| 2990 |
+
"loss": 27.1666,
|
| 2991 |
+
"step": 4260
|
| 2992 |
+
},
|
| 2993 |
+
{
|
| 2994 |
+
"epoch": 0.07923970707288615,
|
| 2995 |
+
"grad_norm": 35.3125,
|
| 2996 |
+
"learning_rate": 9.98761879935989e-06,
|
| 2997 |
+
"loss": 26.8474,
|
| 2998 |
+
"step": 4270
|
| 2999 |
+
},
|
| 3000 |
+
{
|
| 3001 |
+
"epoch": 0.07942528015736598,
|
| 3002 |
+
"grad_norm": 34.65625,
|
| 3003 |
+
"learning_rate": 9.987589803573847e-06,
|
| 3004 |
+
"loss": 27.1325,
|
| 3005 |
+
"step": 4280
|
| 3006 |
+
},
|
| 3007 |
+
{
|
| 3008 |
+
"epoch": 0.0796108532418458,
|
| 3009 |
+
"grad_norm": 33.6875,
|
| 3010 |
+
"learning_rate": 9.987560807787805e-06,
|
| 3011 |
+
"loss": 27.2594,
|
| 3012 |
+
"step": 4290
|
| 3013 |
+
},
|
| 3014 |
+
{
|
| 3015 |
+
"epoch": 0.07979642632632562,
|
| 3016 |
+
"grad_norm": 33.1875,
|
| 3017 |
+
"learning_rate": 9.987531812001762e-06,
|
| 3018 |
+
"loss": 26.8724,
|
| 3019 |
+
"step": 4300
|
| 3020 |
+
},
|
| 3021 |
+
{
|
| 3022 |
+
"epoch": 0.07998199941080546,
|
| 3023 |
+
"grad_norm": 35.375,
|
| 3024 |
+
"learning_rate": 9.987502816215721e-06,
|
| 3025 |
+
"loss": 27.1834,
|
| 3026 |
+
"step": 4310
|
| 3027 |
+
},
|
| 3028 |
+
{
|
| 3029 |
+
"epoch": 0.08016757249528528,
|
| 3030 |
+
"grad_norm": 33.6875,
|
| 3031 |
+
"learning_rate": 9.987473820429677e-06,
|
| 3032 |
+
"loss": 26.8924,
|
| 3033 |
+
"step": 4320
|
| 3034 |
+
},
|
| 3035 |
+
{
|
| 3036 |
+
"epoch": 0.08035314557976511,
|
| 3037 |
+
"grad_norm": 33.90625,
|
| 3038 |
+
"learning_rate": 9.987444824643634e-06,
|
| 3039 |
+
"loss": 26.6646,
|
| 3040 |
+
"step": 4330
|
| 3041 |
+
},
|
| 3042 |
+
{
|
| 3043 |
+
"epoch": 0.08053871866424493,
|
| 3044 |
+
"grad_norm": 32.59375,
|
| 3045 |
+
"learning_rate": 9.987415828857593e-06,
|
| 3046 |
+
"loss": 26.7559,
|
| 3047 |
+
"step": 4340
|
| 3048 |
+
},
|
| 3049 |
+
{
|
| 3050 |
+
"epoch": 0.08072429174872477,
|
| 3051 |
+
"grad_norm": 34.90625,
|
| 3052 |
+
"learning_rate": 9.987386833071551e-06,
|
| 3053 |
+
"loss": 27.3586,
|
| 3054 |
+
"step": 4350
|
| 3055 |
+
},
|
| 3056 |
+
{
|
| 3057 |
+
"epoch": 0.08090986483320459,
|
| 3058 |
+
"grad_norm": 33.90625,
|
| 3059 |
+
"learning_rate": 9.987357837285508e-06,
|
| 3060 |
+
"loss": 26.9466,
|
| 3061 |
+
"step": 4360
|
| 3062 |
+
},
|
| 3063 |
+
{
|
| 3064 |
+
"epoch": 0.08109543791768442,
|
| 3065 |
+
"grad_norm": 35.4375,
|
| 3066 |
+
"learning_rate": 9.987328841499466e-06,
|
| 3067 |
+
"loss": 27.0701,
|
| 3068 |
+
"step": 4370
|
| 3069 |
+
},
|
| 3070 |
+
{
|
| 3071 |
+
"epoch": 0.08128101100216424,
|
| 3072 |
+
"grad_norm": 35.4375,
|
| 3073 |
+
"learning_rate": 9.987299845713423e-06,
|
| 3074 |
+
"loss": 26.6192,
|
| 3075 |
+
"step": 4380
|
| 3076 |
+
},
|
| 3077 |
+
{
|
| 3078 |
+
"epoch": 0.08146658408664408,
|
| 3079 |
+
"grad_norm": 35.34375,
|
| 3080 |
+
"learning_rate": 9.98727084992738e-06,
|
| 3081 |
+
"loss": 26.7397,
|
| 3082 |
+
"step": 4390
|
| 3083 |
+
},
|
| 3084 |
+
{
|
| 3085 |
+
"epoch": 0.0816521571711239,
|
| 3086 |
+
"grad_norm": 33.15625,
|
| 3087 |
+
"learning_rate": 9.987241854141338e-06,
|
| 3088 |
+
"loss": 27.2757,
|
| 3089 |
+
"step": 4400
|
| 3090 |
+
},
|
| 3091 |
+
{
|
| 3092 |
+
"epoch": 0.08183773025560373,
|
| 3093 |
+
"grad_norm": 34.21875,
|
| 3094 |
+
"learning_rate": 9.987212858355297e-06,
|
| 3095 |
+
"loss": 27.0461,
|
| 3096 |
+
"step": 4410
|
| 3097 |
+
},
|
| 3098 |
+
{
|
| 3099 |
+
"epoch": 0.08202330334008355,
|
| 3100 |
+
"grad_norm": 35.15625,
|
| 3101 |
+
"learning_rate": 9.987183862569254e-06,
|
| 3102 |
+
"loss": 26.959,
|
| 3103 |
+
"step": 4420
|
| 3104 |
+
},
|
| 3105 |
+
{
|
| 3106 |
+
"epoch": 0.08220887642456338,
|
| 3107 |
+
"grad_norm": 32.15625,
|
| 3108 |
+
"learning_rate": 9.98715486678321e-06,
|
| 3109 |
+
"loss": 26.739,
|
| 3110 |
+
"step": 4430
|
| 3111 |
+
},
|
| 3112 |
+
{
|
| 3113 |
+
"epoch": 0.0823944495090432,
|
| 3114 |
+
"grad_norm": 32.625,
|
| 3115 |
+
"learning_rate": 9.98712587099717e-06,
|
| 3116 |
+
"loss": 26.9358,
|
| 3117 |
+
"step": 4440
|
| 3118 |
+
},
|
| 3119 |
+
{
|
| 3120 |
+
"epoch": 0.08258002259352304,
|
| 3121 |
+
"grad_norm": 33.6875,
|
| 3122 |
+
"learning_rate": 9.987096875211127e-06,
|
| 3123 |
+
"loss": 26.798,
|
| 3124 |
+
"step": 4450
|
| 3125 |
+
},
|
| 3126 |
+
{
|
| 3127 |
+
"epoch": 0.08276559567800286,
|
| 3128 |
+
"grad_norm": 33.4375,
|
| 3129 |
+
"learning_rate": 9.987067879425084e-06,
|
| 3130 |
+
"loss": 27.0618,
|
| 3131 |
+
"step": 4460
|
| 3132 |
+
},
|
| 3133 |
+
{
|
| 3134 |
+
"epoch": 0.0829511687624827,
|
| 3135 |
+
"grad_norm": 33.15625,
|
| 3136 |
+
"learning_rate": 9.987038883639041e-06,
|
| 3137 |
+
"loss": 26.6068,
|
| 3138 |
+
"step": 4470
|
| 3139 |
+
},
|
| 3140 |
+
{
|
| 3141 |
+
"epoch": 0.08313674184696251,
|
| 3142 |
+
"grad_norm": 34.4375,
|
| 3143 |
+
"learning_rate": 9.987009887852999e-06,
|
| 3144 |
+
"loss": 26.5002,
|
| 3145 |
+
"step": 4480
|
| 3146 |
+
},
|
| 3147 |
+
{
|
| 3148 |
+
"epoch": 0.08332231493144235,
|
| 3149 |
+
"grad_norm": 32.96875,
|
| 3150 |
+
"learning_rate": 9.986980892066956e-06,
|
| 3151 |
+
"loss": 27.0752,
|
| 3152 |
+
"step": 4490
|
| 3153 |
+
},
|
| 3154 |
+
{
|
| 3155 |
+
"epoch": 0.08350788801592217,
|
| 3156 |
+
"grad_norm": 34.625,
|
| 3157 |
+
"learning_rate": 9.986951896280914e-06,
|
| 3158 |
+
"loss": 27.043,
|
| 3159 |
+
"step": 4500
|
| 3160 |
+
},
|
| 3161 |
+
{
|
| 3162 |
+
"epoch": 0.083693461100402,
|
| 3163 |
+
"grad_norm": 32.96875,
|
| 3164 |
+
"learning_rate": 9.986922900494873e-06,
|
| 3165 |
+
"loss": 26.7434,
|
| 3166 |
+
"step": 4510
|
| 3167 |
+
},
|
| 3168 |
+
{
|
| 3169 |
+
"epoch": 0.08387903418488182,
|
| 3170 |
+
"grad_norm": 33.96875,
|
| 3171 |
+
"learning_rate": 9.98689390470883e-06,
|
| 3172 |
+
"loss": 26.9011,
|
| 3173 |
+
"step": 4520
|
| 3174 |
+
},
|
| 3175 |
+
{
|
| 3176 |
+
"epoch": 0.08406460726936166,
|
| 3177 |
+
"grad_norm": 34.75,
|
| 3178 |
+
"learning_rate": 9.986864908922788e-06,
|
| 3179 |
+
"loss": 27.0535,
|
| 3180 |
+
"step": 4530
|
| 3181 |
+
},
|
| 3182 |
+
{
|
| 3183 |
+
"epoch": 0.08425018035384148,
|
| 3184 |
+
"grad_norm": 32.59375,
|
| 3185 |
+
"learning_rate": 9.986835913136745e-06,
|
| 3186 |
+
"loss": 26.9796,
|
| 3187 |
+
"step": 4540
|
| 3188 |
+
},
|
| 3189 |
+
{
|
| 3190 |
+
"epoch": 0.08443575343832131,
|
| 3191 |
+
"grad_norm": 32.71875,
|
| 3192 |
+
"learning_rate": 9.986806917350702e-06,
|
| 3193 |
+
"loss": 26.9757,
|
| 3194 |
+
"step": 4550
|
| 3195 |
+
},
|
| 3196 |
+
{
|
| 3197 |
+
"epoch": 0.08462132652280113,
|
| 3198 |
+
"grad_norm": 33.625,
|
| 3199 |
+
"learning_rate": 9.98677792156466e-06,
|
| 3200 |
+
"loss": 26.684,
|
| 3201 |
+
"step": 4560
|
| 3202 |
+
},
|
| 3203 |
+
{
|
| 3204 |
+
"epoch": 0.08480689960728097,
|
| 3205 |
+
"grad_norm": 33.90625,
|
| 3206 |
+
"learning_rate": 9.986748925778617e-06,
|
| 3207 |
+
"loss": 26.4476,
|
| 3208 |
+
"step": 4570
|
| 3209 |
+
},
|
| 3210 |
+
{
|
| 3211 |
+
"epoch": 0.08499247269176079,
|
| 3212 |
+
"grad_norm": 33.15625,
|
| 3213 |
+
"learning_rate": 9.986719929992575e-06,
|
| 3214 |
+
"loss": 26.6824,
|
| 3215 |
+
"step": 4580
|
| 3216 |
+
},
|
| 3217 |
+
{
|
| 3218 |
+
"epoch": 0.08517804577624061,
|
| 3219 |
+
"grad_norm": 34.78125,
|
| 3220 |
+
"learning_rate": 9.986690934206532e-06,
|
| 3221 |
+
"loss": 26.9472,
|
| 3222 |
+
"step": 4590
|
| 3223 |
+
},
|
| 3224 |
+
{
|
| 3225 |
+
"epoch": 0.08536361886072044,
|
| 3226 |
+
"grad_norm": 34.25,
|
| 3227 |
+
"learning_rate": 9.98666193842049e-06,
|
| 3228 |
+
"loss": 27.1505,
|
| 3229 |
+
"step": 4600
|
| 3230 |
+
},
|
| 3231 |
+
{
|
| 3232 |
+
"epoch": 0.08554919194520026,
|
| 3233 |
+
"grad_norm": 33.1875,
|
| 3234 |
+
"learning_rate": 9.986632942634449e-06,
|
| 3235 |
+
"loss": 26.9236,
|
| 3236 |
+
"step": 4610
|
| 3237 |
+
},
|
| 3238 |
+
{
|
| 3239 |
+
"epoch": 0.0857347650296801,
|
| 3240 |
+
"grad_norm": 34.71875,
|
| 3241 |
+
"learning_rate": 9.986603946848406e-06,
|
| 3242 |
+
"loss": 27.303,
|
| 3243 |
+
"step": 4620
|
| 3244 |
+
},
|
| 3245 |
+
{
|
| 3246 |
+
"epoch": 0.08592033811415992,
|
| 3247 |
+
"grad_norm": 34.09375,
|
| 3248 |
+
"learning_rate": 9.986574951062363e-06,
|
| 3249 |
+
"loss": 26.6816,
|
| 3250 |
+
"step": 4630
|
| 3251 |
+
},
|
| 3252 |
+
{
|
| 3253 |
+
"epoch": 0.08610591119863975,
|
| 3254 |
+
"grad_norm": 34.9375,
|
| 3255 |
+
"learning_rate": 9.98654595527632e-06,
|
| 3256 |
+
"loss": 26.5109,
|
| 3257 |
+
"step": 4640
|
| 3258 |
+
},
|
| 3259 |
+
{
|
| 3260 |
+
"epoch": 0.08629148428311957,
|
| 3261 |
+
"grad_norm": 32.59375,
|
| 3262 |
+
"learning_rate": 9.986516959490278e-06,
|
| 3263 |
+
"loss": 26.4418,
|
| 3264 |
+
"step": 4650
|
| 3265 |
+
},
|
| 3266 |
+
{
|
| 3267 |
+
"epoch": 0.0864770573675994,
|
| 3268 |
+
"grad_norm": 33.8125,
|
| 3269 |
+
"learning_rate": 9.986487963704236e-06,
|
| 3270 |
+
"loss": 26.6267,
|
| 3271 |
+
"step": 4660
|
| 3272 |
+
},
|
| 3273 |
+
{
|
| 3274 |
+
"epoch": 0.08666263045207923,
|
| 3275 |
+
"grad_norm": 33.9375,
|
| 3276 |
+
"learning_rate": 9.986458967918193e-06,
|
| 3277 |
+
"loss": 26.784,
|
| 3278 |
+
"step": 4670
|
| 3279 |
+
},
|
| 3280 |
+
{
|
| 3281 |
+
"epoch": 0.08684820353655906,
|
| 3282 |
+
"grad_norm": 33.78125,
|
| 3283 |
+
"learning_rate": 9.98642997213215e-06,
|
| 3284 |
+
"loss": 26.6958,
|
| 3285 |
+
"step": 4680
|
| 3286 |
+
},
|
| 3287 |
+
{
|
| 3288 |
+
"epoch": 0.08703377662103888,
|
| 3289 |
+
"grad_norm": 33.4375,
|
| 3290 |
+
"learning_rate": 9.98640097634611e-06,
|
| 3291 |
+
"loss": 26.3625,
|
| 3292 |
+
"step": 4690
|
| 3293 |
+
},
|
| 3294 |
+
{
|
| 3295 |
+
"epoch": 0.08721934970551871,
|
| 3296 |
+
"grad_norm": 31.953125,
|
| 3297 |
+
"learning_rate": 9.986371980560065e-06,
|
| 3298 |
+
"loss": 26.9694,
|
| 3299 |
+
"step": 4700
|
| 3300 |
+
},
|
| 3301 |
+
{
|
| 3302 |
+
"epoch": 0.08740492278999853,
|
| 3303 |
+
"grad_norm": 33.59375,
|
| 3304 |
+
"learning_rate": 9.986342984774023e-06,
|
| 3305 |
+
"loss": 26.4038,
|
| 3306 |
+
"step": 4710
|
| 3307 |
+
},
|
| 3308 |
+
{
|
| 3309 |
+
"epoch": 0.08759049587447837,
|
| 3310 |
+
"grad_norm": 32.75,
|
| 3311 |
+
"learning_rate": 9.986313988987982e-06,
|
| 3312 |
+
"loss": 26.7672,
|
| 3313 |
+
"step": 4720
|
| 3314 |
+
},
|
| 3315 |
+
{
|
| 3316 |
+
"epoch": 0.08777606895895819,
|
| 3317 |
+
"grad_norm": 34.375,
|
| 3318 |
+
"learning_rate": 9.986284993201939e-06,
|
| 3319 |
+
"loss": 26.8111,
|
| 3320 |
+
"step": 4730
|
| 3321 |
+
},
|
| 3322 |
+
{
|
| 3323 |
+
"epoch": 0.08796164204343802,
|
| 3324 |
+
"grad_norm": 34.0625,
|
| 3325 |
+
"learning_rate": 9.986255997415897e-06,
|
| 3326 |
+
"loss": 26.6389,
|
| 3327 |
+
"step": 4740
|
| 3328 |
+
},
|
| 3329 |
+
{
|
| 3330 |
+
"epoch": 0.08814721512791784,
|
| 3331 |
+
"grad_norm": 34.4375,
|
| 3332 |
+
"learning_rate": 9.986227001629854e-06,
|
| 3333 |
+
"loss": 26.8616,
|
| 3334 |
+
"step": 4750
|
| 3335 |
+
},
|
| 3336 |
+
{
|
| 3337 |
+
"epoch": 0.08833278821239768,
|
| 3338 |
+
"grad_norm": 33.34375,
|
| 3339 |
+
"learning_rate": 9.986198005843811e-06,
|
| 3340 |
+
"loss": 26.621,
|
| 3341 |
+
"step": 4760
|
| 3342 |
+
},
|
| 3343 |
+
{
|
| 3344 |
+
"epoch": 0.0885183612968775,
|
| 3345 |
+
"grad_norm": 31.96875,
|
| 3346 |
+
"learning_rate": 9.986169010057769e-06,
|
| 3347 |
+
"loss": 26.5185,
|
| 3348 |
+
"step": 4770
|
| 3349 |
+
},
|
| 3350 |
+
{
|
| 3351 |
+
"epoch": 0.08870393438135733,
|
| 3352 |
+
"grad_norm": 32.5625,
|
| 3353 |
+
"learning_rate": 9.986140014271726e-06,
|
| 3354 |
+
"loss": 26.8874,
|
| 3355 |
+
"step": 4780
|
| 3356 |
+
},
|
| 3357 |
+
{
|
| 3358 |
+
"epoch": 0.08888950746583715,
|
| 3359 |
+
"grad_norm": 34.03125,
|
| 3360 |
+
"learning_rate": 9.986111018485685e-06,
|
| 3361 |
+
"loss": 26.7579,
|
| 3362 |
+
"step": 4790
|
| 3363 |
+
},
|
| 3364 |
+
{
|
| 3365 |
+
"epoch": 0.08907508055031699,
|
| 3366 |
+
"grad_norm": 35.96875,
|
| 3367 |
+
"learning_rate": 9.986082022699641e-06,
|
| 3368 |
+
"loss": 26.8933,
|
| 3369 |
+
"step": 4800
|
| 3370 |
+
},
|
| 3371 |
+
{
|
| 3372 |
+
"epoch": 0.08926065363479681,
|
| 3373 |
+
"grad_norm": 35.03125,
|
| 3374 |
+
"learning_rate": 9.986053026913598e-06,
|
| 3375 |
+
"loss": 26.5203,
|
| 3376 |
+
"step": 4810
|
| 3377 |
+
},
|
| 3378 |
+
{
|
| 3379 |
+
"epoch": 0.08944622671927664,
|
| 3380 |
+
"grad_norm": 35.1875,
|
| 3381 |
+
"learning_rate": 9.986024031127557e-06,
|
| 3382 |
+
"loss": 26.714,
|
| 3383 |
+
"step": 4820
|
| 3384 |
+
},
|
| 3385 |
+
{
|
| 3386 |
+
"epoch": 0.08963179980375646,
|
| 3387 |
+
"grad_norm": 34.78125,
|
| 3388 |
+
"learning_rate": 9.985995035341515e-06,
|
| 3389 |
+
"loss": 26.6803,
|
| 3390 |
+
"step": 4830
|
| 3391 |
+
},
|
| 3392 |
+
{
|
| 3393 |
+
"epoch": 0.0898173728882363,
|
| 3394 |
+
"grad_norm": 37.03125,
|
| 3395 |
+
"learning_rate": 9.985966039555472e-06,
|
| 3396 |
+
"loss": 26.7259,
|
| 3397 |
+
"step": 4840
|
| 3398 |
+
},
|
| 3399 |
+
{
|
| 3400 |
+
"epoch": 0.09000294597271612,
|
| 3401 |
+
"grad_norm": 34.28125,
|
| 3402 |
+
"learning_rate": 9.98593704376943e-06,
|
| 3403 |
+
"loss": 27.119,
|
| 3404 |
+
"step": 4850
|
| 3405 |
+
},
|
| 3406 |
+
{
|
| 3407 |
+
"epoch": 0.09018851905719595,
|
| 3408 |
+
"grad_norm": 33.65625,
|
| 3409 |
+
"learning_rate": 9.985908047983387e-06,
|
| 3410 |
+
"loss": 26.4196,
|
| 3411 |
+
"step": 4860
|
| 3412 |
+
},
|
| 3413 |
+
{
|
| 3414 |
+
"epoch": 0.09037409214167577,
|
| 3415 |
+
"grad_norm": 33.84375,
|
| 3416 |
+
"learning_rate": 9.985879052197344e-06,
|
| 3417 |
+
"loss": 26.3284,
|
| 3418 |
+
"step": 4870
|
| 3419 |
+
},
|
| 3420 |
+
{
|
| 3421 |
+
"epoch": 0.09055966522615559,
|
| 3422 |
+
"grad_norm": 32.84375,
|
| 3423 |
+
"learning_rate": 9.985850056411302e-06,
|
| 3424 |
+
"loss": 26.8074,
|
| 3425 |
+
"step": 4880
|
| 3426 |
+
},
|
| 3427 |
+
{
|
| 3428 |
+
"epoch": 0.09074523831063543,
|
| 3429 |
+
"grad_norm": 34.15625,
|
| 3430 |
+
"learning_rate": 9.985821060625261e-06,
|
| 3431 |
+
"loss": 26.4722,
|
| 3432 |
+
"step": 4890
|
| 3433 |
+
},
|
| 3434 |
+
{
|
| 3435 |
+
"epoch": 0.09093081139511525,
|
| 3436 |
+
"grad_norm": 33.125,
|
| 3437 |
+
"learning_rate": 9.985792064839218e-06,
|
| 3438 |
+
"loss": 26.3422,
|
| 3439 |
+
"step": 4900
|
| 3440 |
+
},
|
| 3441 |
+
{
|
| 3442 |
+
"epoch": 0.09111638447959508,
|
| 3443 |
+
"grad_norm": 33.75,
|
| 3444 |
+
"learning_rate": 9.985763069053174e-06,
|
| 3445 |
+
"loss": 26.2426,
|
| 3446 |
+
"step": 4910
|
| 3447 |
+
},
|
| 3448 |
+
{
|
| 3449 |
+
"epoch": 0.0913019575640749,
|
| 3450 |
+
"grad_norm": 34.1875,
|
| 3451 |
+
"learning_rate": 9.985734073267133e-06,
|
| 3452 |
+
"loss": 26.4652,
|
| 3453 |
+
"step": 4920
|
| 3454 |
+
},
|
| 3455 |
+
{
|
| 3456 |
+
"epoch": 0.09148753064855474,
|
| 3457 |
+
"grad_norm": 34.125,
|
| 3458 |
+
"learning_rate": 9.98570507748109e-06,
|
| 3459 |
+
"loss": 26.6846,
|
| 3460 |
+
"step": 4930
|
| 3461 |
+
},
|
| 3462 |
+
{
|
| 3463 |
+
"epoch": 0.09167310373303456,
|
| 3464 |
+
"grad_norm": 32.15625,
|
| 3465 |
+
"learning_rate": 9.985676081695048e-06,
|
| 3466 |
+
"loss": 26.6221,
|
| 3467 |
+
"step": 4940
|
| 3468 |
+
},
|
| 3469 |
+
{
|
| 3470 |
+
"epoch": 0.09185867681751439,
|
| 3471 |
+
"grad_norm": 32.75,
|
| 3472 |
+
"learning_rate": 9.985647085909005e-06,
|
| 3473 |
+
"loss": 26.6039,
|
| 3474 |
+
"step": 4950
|
| 3475 |
+
},
|
| 3476 |
+
{
|
| 3477 |
+
"epoch": 0.09204424990199421,
|
| 3478 |
+
"grad_norm": 34.15625,
|
| 3479 |
+
"learning_rate": 9.985618090122965e-06,
|
| 3480 |
+
"loss": 26.6058,
|
| 3481 |
+
"step": 4960
|
| 3482 |
+
},
|
| 3483 |
+
{
|
| 3484 |
+
"epoch": 0.09222982298647404,
|
| 3485 |
+
"grad_norm": 34.75,
|
| 3486 |
+
"learning_rate": 9.98558909433692e-06,
|
| 3487 |
+
"loss": 26.4243,
|
| 3488 |
+
"step": 4970
|
| 3489 |
+
},
|
| 3490 |
+
{
|
| 3491 |
+
"epoch": 0.09241539607095386,
|
| 3492 |
+
"grad_norm": 35.875,
|
| 3493 |
+
"learning_rate": 9.985560098550878e-06,
|
| 3494 |
+
"loss": 27.0457,
|
| 3495 |
+
"step": 4980
|
| 3496 |
+
},
|
| 3497 |
+
{
|
| 3498 |
+
"epoch": 0.0926009691554337,
|
| 3499 |
+
"grad_norm": 34.75,
|
| 3500 |
+
"learning_rate": 9.985531102764837e-06,
|
| 3501 |
+
"loss": 26.3863,
|
| 3502 |
+
"step": 4990
|
| 3503 |
+
},
|
| 3504 |
+
{
|
| 3505 |
+
"epoch": 0.09278654223991352,
|
| 3506 |
+
"grad_norm": 35.3125,
|
| 3507 |
+
"learning_rate": 9.985502106978794e-06,
|
| 3508 |
+
"loss": 26.6178,
|
| 3509 |
+
"step": 5000
|
| 3510 |
+
},
|
| 3511 |
+
{
|
| 3512 |
+
"epoch": 0.09278654223991352,
|
| 3513 |
+
"eval_loss": 3.30993914604187,
|
| 3514 |
+
"eval_runtime": 453.9307,
|
| 3515 |
+
"eval_samples_per_second": 3198.984,
|
| 3516 |
+
"eval_steps_per_second": 49.986,
|
| 3517 |
+
"step": 5000
|
| 3518 |
+
}
|
| 3519 |
+
],
|
| 3520 |
+
"logging_steps": 10,
|
| 3521 |
+
"max_steps": 53887,
|
| 3522 |
+
"num_input_tokens_seen": 0,
|
| 3523 |
+
"num_train_epochs": 1,
|
| 3524 |
+
"save_steps": 5000,
|
| 3525 |
+
"stateful_callbacks": {
|
| 3526 |
+
"TrainerControl": {
|
| 3527 |
+
"args": {
|
| 3528 |
+
"should_epoch_stop": false,
|
| 3529 |
+
"should_evaluate": false,
|
| 3530 |
+
"should_log": false,
|
| 3531 |
+
"should_save": true,
|
| 3532 |
+
"should_training_stop": false
|
| 3533 |
+
},
|
| 3534 |
+
"attributes": {}
|
| 3535 |
+
}
|
| 3536 |
+
},
|
| 3537 |
+
"total_flos": 8.72761379520512e+17,
|
| 3538 |
+
"train_batch_size": 8,
|
| 3539 |
+
"trial_name": null,
|
| 3540 |
+
"trial_params": null
|
| 3541 |
+
}
|
last-checkpoint/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8d0cbc0eed40136cd693c3a06f37f23f3d1f279fb4b9a64d5a0e7c17f266fe16
|
| 3 |
+
size 5368
|