🚀 Upload Savant–Neo Φ9.4 hybrid model
Browse files- checkpoint-7500/config.json +27 -0
- checkpoint-7500/generation_config.json +5 -0
- checkpoint-7500/model.safetensors +3 -0
- checkpoint-7500/optimizer.pt +3 -0
- checkpoint-7500/rng_state.pth +3 -0
- checkpoint-7500/scaler.pt +3 -0
- checkpoint-7500/scheduler.pt +3 -0
- checkpoint-7500/trainer_state.json +2134 -0
- checkpoint-7500/training_args.bin +3 -0
- checkpoint-7906/config.json +27 -0
- checkpoint-7906/generation_config.json +5 -0
- checkpoint-7906/model.safetensors +3 -0
- checkpoint-7906/optimizer.pt +3 -0
- checkpoint-7906/rng_state.pth +3 -0
- checkpoint-7906/scaler.pt +3 -0
- checkpoint-7906/scheduler.pt +3 -0
- checkpoint-7906/trainer_state.json +2246 -0
- checkpoint-7906/training_args.bin +3 -0
- config.json +27 -0
- generation_config.json +5 -0
- model.safetensors +3 -0
- special_tokens_map.json +37 -0
- tokenizer_config.json +65 -0
- vocab.txt +0 -0
checkpoint-7500/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_cross_attention": true,
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertLMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"gradient_checkpointing": false,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 384,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 1536,
|
| 15 |
+
"is_decoder": true,
|
| 16 |
+
"layer_norm_eps": 1e-12,
|
| 17 |
+
"max_position_embeddings": 512,
|
| 18 |
+
"model_type": "bert",
|
| 19 |
+
"num_attention_heads": 12,
|
| 20 |
+
"num_hidden_layers": 6,
|
| 21 |
+
"pad_token_id": 0,
|
| 22 |
+
"position_embedding_type": "absolute",
|
| 23 |
+
"transformers_version": "4.57.1",
|
| 24 |
+
"type_vocab_size": 2,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 30522
|
| 27 |
+
}
|
checkpoint-7500/generation_config.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"pad_token_id": 0,
|
| 4 |
+
"transformers_version": "4.57.1"
|
| 5 |
+
}
|
checkpoint-7500/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a2bf5efaffc5881b293cd562b32fcbbff88eb4f87a27492838d86ed533c6fcde
|
| 3 |
+
size 90990200
|
checkpoint-7500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a85e5e3d1dc5fb77172c22504cfe7a1d59adc7d07dc7b46c5790257e2042e019
|
| 3 |
+
size 182045763
|
checkpoint-7500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:414f7111173bfce721fed26faa806dd108f69d7b7fa3e19686cef203983939a6
|
| 3 |
+
size 14565
|
checkpoint-7500/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cc99cd17a64f2282f11fb8d0496be9bbede48f585c93fecdd6c50a4718dd4592
|
| 3 |
+
size 1389
|
checkpoint-7500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:07e85f6df6c35a6bff7c605ed8b90329f8480f18c29d4665ee1a1dfdf3214a2b
|
| 3 |
+
size 1453
|
checkpoint-7500/trainer_state.json
ADDED
|
@@ -0,0 +1,2134 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.9486465975208702,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 7500,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0031621553250695674,
|
| 14 |
+
"grad_norm": 11.55327320098877,
|
| 15 |
+
"learning_rate": 1.9939286617758668e-05,
|
| 16 |
+
"loss": 8.596,
|
| 17 |
+
"step": 25
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.006324310650139135,
|
| 21 |
+
"grad_norm": 5.928526401519775,
|
| 22 |
+
"learning_rate": 1.9876043511257274e-05,
|
| 23 |
+
"loss": 5.501,
|
| 24 |
+
"step": 50
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.009486465975208701,
|
| 28 |
+
"grad_norm": 2.7216451168060303,
|
| 29 |
+
"learning_rate": 1.9812800404755884e-05,
|
| 30 |
+
"loss": 4.9742,
|
| 31 |
+
"step": 75
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.01264862130027827,
|
| 35 |
+
"grad_norm": 2.725499153137207,
|
| 36 |
+
"learning_rate": 1.974955729825449e-05,
|
| 37 |
+
"loss": 4.6023,
|
| 38 |
+
"step": 100
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.015810776625347838,
|
| 42 |
+
"grad_norm": 2.318068027496338,
|
| 43 |
+
"learning_rate": 1.96863141917531e-05,
|
| 44 |
+
"loss": 4.4124,
|
| 45 |
+
"step": 125
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.018972931950417403,
|
| 49 |
+
"grad_norm": 2.437415838241577,
|
| 50 |
+
"learning_rate": 1.9623071085251707e-05,
|
| 51 |
+
"loss": 4.5471,
|
| 52 |
+
"step": 150
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.02213508727548697,
|
| 56 |
+
"grad_norm": 1.6126755475997925,
|
| 57 |
+
"learning_rate": 1.955982797875032e-05,
|
| 58 |
+
"loss": 4.0951,
|
| 59 |
+
"step": 175
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.02529724260055654,
|
| 63 |
+
"grad_norm": 1.9190938472747803,
|
| 64 |
+
"learning_rate": 1.9496584872248927e-05,
|
| 65 |
+
"loss": 4.0572,
|
| 66 |
+
"step": 200
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.028459397925626108,
|
| 70 |
+
"grad_norm": 1.6798152923583984,
|
| 71 |
+
"learning_rate": 1.9433341765747537e-05,
|
| 72 |
+
"loss": 3.9318,
|
| 73 |
+
"step": 225
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.031621553250695676,
|
| 77 |
+
"grad_norm": 2.02826189994812,
|
| 78 |
+
"learning_rate": 1.9370098659246143e-05,
|
| 79 |
+
"loss": 3.9254,
|
| 80 |
+
"step": 250
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.034783708575765244,
|
| 84 |
+
"grad_norm": 1.6234019994735718,
|
| 85 |
+
"learning_rate": 1.9306855552744753e-05,
|
| 86 |
+
"loss": 3.7144,
|
| 87 |
+
"step": 275
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.037945863900834806,
|
| 91 |
+
"grad_norm": 2.1222445964813232,
|
| 92 |
+
"learning_rate": 1.924361244624336e-05,
|
| 93 |
+
"loss": 4.0748,
|
| 94 |
+
"step": 300
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.041108019225904374,
|
| 98 |
+
"grad_norm": 1.8071123361587524,
|
| 99 |
+
"learning_rate": 1.918036933974197e-05,
|
| 100 |
+
"loss": 3.8597,
|
| 101 |
+
"step": 325
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.04427017455097394,
|
| 105 |
+
"grad_norm": 1.2955825328826904,
|
| 106 |
+
"learning_rate": 1.911712623324058e-05,
|
| 107 |
+
"loss": 3.6891,
|
| 108 |
+
"step": 350
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.04743232987604351,
|
| 112 |
+
"grad_norm": 1.9034608602523804,
|
| 113 |
+
"learning_rate": 1.905388312673919e-05,
|
| 114 |
+
"loss": 3.8955,
|
| 115 |
+
"step": 375
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.05059448520111308,
|
| 119 |
+
"grad_norm": 1.649466872215271,
|
| 120 |
+
"learning_rate": 1.8990640020237796e-05,
|
| 121 |
+
"loss": 3.8186,
|
| 122 |
+
"step": 400
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.05375664052618265,
|
| 126 |
+
"grad_norm": 1.8459181785583496,
|
| 127 |
+
"learning_rate": 1.8927396913736406e-05,
|
| 128 |
+
"loss": 3.5247,
|
| 129 |
+
"step": 425
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.056918795851252216,
|
| 133 |
+
"grad_norm": 1.5839076042175293,
|
| 134 |
+
"learning_rate": 1.8864153807235012e-05,
|
| 135 |
+
"loss": 3.756,
|
| 136 |
+
"step": 450
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.060080951176321784,
|
| 140 |
+
"grad_norm": 1.5614947080612183,
|
| 141 |
+
"learning_rate": 1.8800910700733622e-05,
|
| 142 |
+
"loss": 3.6433,
|
| 143 |
+
"step": 475
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.06324310650139135,
|
| 147 |
+
"grad_norm": 1.7831311225891113,
|
| 148 |
+
"learning_rate": 1.873766759423223e-05,
|
| 149 |
+
"loss": 3.9025,
|
| 150 |
+
"step": 500
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.06640526182646092,
|
| 154 |
+
"grad_norm": 1.4928420782089233,
|
| 155 |
+
"learning_rate": 1.867442448773084e-05,
|
| 156 |
+
"loss": 3.4216,
|
| 157 |
+
"step": 525
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.06956741715153049,
|
| 161 |
+
"grad_norm": 1.8199015855789185,
|
| 162 |
+
"learning_rate": 1.861118138122945e-05,
|
| 163 |
+
"loss": 3.7497,
|
| 164 |
+
"step": 550
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.07272957247660006,
|
| 168 |
+
"grad_norm": 1.6205722093582153,
|
| 169 |
+
"learning_rate": 1.854793827472806e-05,
|
| 170 |
+
"loss": 3.4153,
|
| 171 |
+
"step": 575
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.07589172780166961,
|
| 175 |
+
"grad_norm": 1.4476927518844604,
|
| 176 |
+
"learning_rate": 1.8484695168226665e-05,
|
| 177 |
+
"loss": 3.6298,
|
| 178 |
+
"step": 600
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.07905388312673918,
|
| 182 |
+
"grad_norm": 1.5358431339263916,
|
| 183 |
+
"learning_rate": 1.8421452061725275e-05,
|
| 184 |
+
"loss": 3.5282,
|
| 185 |
+
"step": 625
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.08221603845180875,
|
| 189 |
+
"grad_norm": 1.616495966911316,
|
| 190 |
+
"learning_rate": 1.835820895522388e-05,
|
| 191 |
+
"loss": 3.6029,
|
| 192 |
+
"step": 650
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.08537819377687832,
|
| 196 |
+
"grad_norm": 1.4835790395736694,
|
| 197 |
+
"learning_rate": 1.829496584872249e-05,
|
| 198 |
+
"loss": 3.6817,
|
| 199 |
+
"step": 675
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.08854034910194788,
|
| 203 |
+
"grad_norm": 1.8951404094696045,
|
| 204 |
+
"learning_rate": 1.8231722742221098e-05,
|
| 205 |
+
"loss": 3.5222,
|
| 206 |
+
"step": 700
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.09170250442701745,
|
| 210 |
+
"grad_norm": 1.4792296886444092,
|
| 211 |
+
"learning_rate": 1.8168479635719707e-05,
|
| 212 |
+
"loss": 3.552,
|
| 213 |
+
"step": 725
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.09486465975208702,
|
| 217 |
+
"grad_norm": 1.5331674814224243,
|
| 218 |
+
"learning_rate": 1.8105236529218317e-05,
|
| 219 |
+
"loss": 3.6274,
|
| 220 |
+
"step": 750
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.09802681507715659,
|
| 224 |
+
"grad_norm": 1.3521767854690552,
|
| 225 |
+
"learning_rate": 1.8041993422716927e-05,
|
| 226 |
+
"loss": 3.4604,
|
| 227 |
+
"step": 775
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.10118897040222616,
|
| 231 |
+
"grad_norm": 1.9199748039245605,
|
| 232 |
+
"learning_rate": 1.7978750316215534e-05,
|
| 233 |
+
"loss": 3.6434,
|
| 234 |
+
"step": 800
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.10435112572729573,
|
| 238 |
+
"grad_norm": 1.4599398374557495,
|
| 239 |
+
"learning_rate": 1.7915507209714144e-05,
|
| 240 |
+
"loss": 3.4381,
|
| 241 |
+
"step": 825
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.1075132810523653,
|
| 245 |
+
"grad_norm": 1.8845282793045044,
|
| 246 |
+
"learning_rate": 1.785226410321275e-05,
|
| 247 |
+
"loss": 3.4885,
|
| 248 |
+
"step": 850
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.11067543637743486,
|
| 252 |
+
"grad_norm": 1.6054080724716187,
|
| 253 |
+
"learning_rate": 1.778902099671136e-05,
|
| 254 |
+
"loss": 3.4714,
|
| 255 |
+
"step": 875
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.11383759170250443,
|
| 259 |
+
"grad_norm": 1.4243124723434448,
|
| 260 |
+
"learning_rate": 1.7725777890209967e-05,
|
| 261 |
+
"loss": 3.423,
|
| 262 |
+
"step": 900
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.116999747027574,
|
| 266 |
+
"grad_norm": 1.828045129776001,
|
| 267 |
+
"learning_rate": 1.7662534783708576e-05,
|
| 268 |
+
"loss": 3.3451,
|
| 269 |
+
"step": 925
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.12016190235264357,
|
| 273 |
+
"grad_norm": 1.7184784412384033,
|
| 274 |
+
"learning_rate": 1.7599291677207186e-05,
|
| 275 |
+
"loss": 3.4567,
|
| 276 |
+
"step": 950
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.12332405767771312,
|
| 280 |
+
"grad_norm": 2.0031352043151855,
|
| 281 |
+
"learning_rate": 1.7536048570705796e-05,
|
| 282 |
+
"loss": 3.5227,
|
| 283 |
+
"step": 975
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.1264862130027827,
|
| 287 |
+
"grad_norm": 1.8211214542388916,
|
| 288 |
+
"learning_rate": 1.7472805464204403e-05,
|
| 289 |
+
"loss": 3.4222,
|
| 290 |
+
"step": 1000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.12964836832785226,
|
| 294 |
+
"grad_norm": 1.6366448402404785,
|
| 295 |
+
"learning_rate": 1.7409562357703013e-05,
|
| 296 |
+
"loss": 3.3479,
|
| 297 |
+
"step": 1025
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.13281052365292184,
|
| 301 |
+
"grad_norm": 1.4770481586456299,
|
| 302 |
+
"learning_rate": 1.734631925120162e-05,
|
| 303 |
+
"loss": 3.4232,
|
| 304 |
+
"step": 1050
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.1359726789779914,
|
| 308 |
+
"grad_norm": 1.3028419017791748,
|
| 309 |
+
"learning_rate": 1.728307614470023e-05,
|
| 310 |
+
"loss": 3.2511,
|
| 311 |
+
"step": 1075
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.13913483430306098,
|
| 315 |
+
"grad_norm": 1.8425815105438232,
|
| 316 |
+
"learning_rate": 1.721983303819884e-05,
|
| 317 |
+
"loss": 3.3851,
|
| 318 |
+
"step": 1100
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.14229698962813053,
|
| 322 |
+
"grad_norm": 1.5002127885818481,
|
| 323 |
+
"learning_rate": 1.715658993169745e-05,
|
| 324 |
+
"loss": 3.5281,
|
| 325 |
+
"step": 1125
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.14545914495320011,
|
| 329 |
+
"grad_norm": 1.5222417116165161,
|
| 330 |
+
"learning_rate": 1.7093346825196055e-05,
|
| 331 |
+
"loss": 3.3473,
|
| 332 |
+
"step": 1150
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.14862130027826967,
|
| 336 |
+
"grad_norm": 1.628788948059082,
|
| 337 |
+
"learning_rate": 1.7030103718694665e-05,
|
| 338 |
+
"loss": 3.3795,
|
| 339 |
+
"step": 1175
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.15178345560333922,
|
| 343 |
+
"grad_norm": 1.3158140182495117,
|
| 344 |
+
"learning_rate": 1.6966860612193272e-05,
|
| 345 |
+
"loss": 3.5536,
|
| 346 |
+
"step": 1200
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.1549456109284088,
|
| 350 |
+
"grad_norm": 1.8109098672866821,
|
| 351 |
+
"learning_rate": 1.690361750569188e-05,
|
| 352 |
+
"loss": 3.3006,
|
| 353 |
+
"step": 1225
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.15810776625347836,
|
| 357 |
+
"grad_norm": 2.0179309844970703,
|
| 358 |
+
"learning_rate": 1.6840374399190488e-05,
|
| 359 |
+
"loss": 3.4737,
|
| 360 |
+
"step": 1250
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.16126992157854794,
|
| 364 |
+
"grad_norm": 1.8066917657852173,
|
| 365 |
+
"learning_rate": 1.6777131292689098e-05,
|
| 366 |
+
"loss": 3.3135,
|
| 367 |
+
"step": 1275
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.1644320769036175,
|
| 371 |
+
"grad_norm": 1.4508979320526123,
|
| 372 |
+
"learning_rate": 1.6713888186187708e-05,
|
| 373 |
+
"loss": 3.154,
|
| 374 |
+
"step": 1300
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.16759423222868708,
|
| 378 |
+
"grad_norm": 1.324156641960144,
|
| 379 |
+
"learning_rate": 1.6650645079686318e-05,
|
| 380 |
+
"loss": 3.3076,
|
| 381 |
+
"step": 1325
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.17075638755375663,
|
| 385 |
+
"grad_norm": 1.7277374267578125,
|
| 386 |
+
"learning_rate": 1.6587401973184924e-05,
|
| 387 |
+
"loss": 3.3168,
|
| 388 |
+
"step": 1350
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.17391854287882622,
|
| 392 |
+
"grad_norm": 1.581597924232483,
|
| 393 |
+
"learning_rate": 1.6524158866683534e-05,
|
| 394 |
+
"loss": 3.2586,
|
| 395 |
+
"step": 1375
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.17708069820389577,
|
| 399 |
+
"grad_norm": 1.2408771514892578,
|
| 400 |
+
"learning_rate": 1.646091576018214e-05,
|
| 401 |
+
"loss": 3.3174,
|
| 402 |
+
"step": 1400
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.18024285352896535,
|
| 406 |
+
"grad_norm": 1.5069456100463867,
|
| 407 |
+
"learning_rate": 1.639767265368075e-05,
|
| 408 |
+
"loss": 3.2248,
|
| 409 |
+
"step": 1425
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.1834050088540349,
|
| 413 |
+
"grad_norm": 1.474717378616333,
|
| 414 |
+
"learning_rate": 1.6334429547179357e-05,
|
| 415 |
+
"loss": 3.2757,
|
| 416 |
+
"step": 1450
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.1865671641791045,
|
| 420 |
+
"grad_norm": 1.648443341255188,
|
| 421 |
+
"learning_rate": 1.6271186440677967e-05,
|
| 422 |
+
"loss": 3.2754,
|
| 423 |
+
"step": 1475
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.18972931950417404,
|
| 427 |
+
"grad_norm": 1.5270763635635376,
|
| 428 |
+
"learning_rate": 1.6207943334176577e-05,
|
| 429 |
+
"loss": 3.2413,
|
| 430 |
+
"step": 1500
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.19289147482924363,
|
| 434 |
+
"grad_norm": 1.643730640411377,
|
| 435 |
+
"learning_rate": 1.6144700227675187e-05,
|
| 436 |
+
"loss": 3.213,
|
| 437 |
+
"step": 1525
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.19605363015431318,
|
| 441 |
+
"grad_norm": 1.4000329971313477,
|
| 442 |
+
"learning_rate": 1.6081457121173793e-05,
|
| 443 |
+
"loss": 3.236,
|
| 444 |
+
"step": 1550
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.19921578547938273,
|
| 448 |
+
"grad_norm": 1.6957166194915771,
|
| 449 |
+
"learning_rate": 1.6018214014672403e-05,
|
| 450 |
+
"loss": 3.1811,
|
| 451 |
+
"step": 1575
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.20237794080445232,
|
| 455 |
+
"grad_norm": 1.763476848602295,
|
| 456 |
+
"learning_rate": 1.595497090817101e-05,
|
| 457 |
+
"loss": 3.2634,
|
| 458 |
+
"step": 1600
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.20554009612952187,
|
| 462 |
+
"grad_norm": 1.5622406005859375,
|
| 463 |
+
"learning_rate": 1.589172780166962e-05,
|
| 464 |
+
"loss": 3.2498,
|
| 465 |
+
"step": 1625
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.20870225145459145,
|
| 469 |
+
"grad_norm": 2.501917600631714,
|
| 470 |
+
"learning_rate": 1.5828484695168226e-05,
|
| 471 |
+
"loss": 3.3526,
|
| 472 |
+
"step": 1650
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.211864406779661,
|
| 476 |
+
"grad_norm": 1.5445410013198853,
|
| 477 |
+
"learning_rate": 1.5765241588666836e-05,
|
| 478 |
+
"loss": 3.25,
|
| 479 |
+
"step": 1675
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.2150265621047306,
|
| 483 |
+
"grad_norm": 1.420432209968567,
|
| 484 |
+
"learning_rate": 1.5701998482165446e-05,
|
| 485 |
+
"loss": 3.316,
|
| 486 |
+
"step": 1700
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.21818871742980014,
|
| 490 |
+
"grad_norm": 1.7002512216567993,
|
| 491 |
+
"learning_rate": 1.5638755375664056e-05,
|
| 492 |
+
"loss": 3.156,
|
| 493 |
+
"step": 1725
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.22135087275486973,
|
| 497 |
+
"grad_norm": 1.533703088760376,
|
| 498 |
+
"learning_rate": 1.5575512269162662e-05,
|
| 499 |
+
"loss": 3.0938,
|
| 500 |
+
"step": 1750
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.22451302807993928,
|
| 504 |
+
"grad_norm": 1.4791340827941895,
|
| 505 |
+
"learning_rate": 1.5512269162661272e-05,
|
| 506 |
+
"loss": 3.2756,
|
| 507 |
+
"step": 1775
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.22767518340500886,
|
| 511 |
+
"grad_norm": 1.783389925956726,
|
| 512 |
+
"learning_rate": 1.544902605615988e-05,
|
| 513 |
+
"loss": 3.3588,
|
| 514 |
+
"step": 1800
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.23083733873007842,
|
| 518 |
+
"grad_norm": 1.7652767896652222,
|
| 519 |
+
"learning_rate": 1.538578294965849e-05,
|
| 520 |
+
"loss": 3.1746,
|
| 521 |
+
"step": 1825
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.233999494055148,
|
| 525 |
+
"grad_norm": 1.7373157739639282,
|
| 526 |
+
"learning_rate": 1.53225398431571e-05,
|
| 527 |
+
"loss": 3.2757,
|
| 528 |
+
"step": 1850
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.23716164938021755,
|
| 532 |
+
"grad_norm": 1.655953049659729,
|
| 533 |
+
"learning_rate": 1.5259296736655705e-05,
|
| 534 |
+
"loss": 3.2366,
|
| 535 |
+
"step": 1875
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.24032380470528714,
|
| 539 |
+
"grad_norm": 1.4808825254440308,
|
| 540 |
+
"learning_rate": 1.5196053630154315e-05,
|
| 541 |
+
"loss": 3.2056,
|
| 542 |
+
"step": 1900
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.2434859600303567,
|
| 546 |
+
"grad_norm": 1.3337376117706299,
|
| 547 |
+
"learning_rate": 1.5132810523652923e-05,
|
| 548 |
+
"loss": 2.9247,
|
| 549 |
+
"step": 1925
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.24664811535542625,
|
| 553 |
+
"grad_norm": 1.3115172386169434,
|
| 554 |
+
"learning_rate": 1.5069567417151531e-05,
|
| 555 |
+
"loss": 3.3964,
|
| 556 |
+
"step": 1950
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.24981027068049583,
|
| 560 |
+
"grad_norm": 1.3567864894866943,
|
| 561 |
+
"learning_rate": 1.500632431065014e-05,
|
| 562 |
+
"loss": 3.2777,
|
| 563 |
+
"step": 1975
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.2529724260055654,
|
| 567 |
+
"grad_norm": 1.5610828399658203,
|
| 568 |
+
"learning_rate": 1.4943081204148748e-05,
|
| 569 |
+
"loss": 3.0115,
|
| 570 |
+
"step": 2000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.25613458133063494,
|
| 574 |
+
"grad_norm": 1.6952314376831055,
|
| 575 |
+
"learning_rate": 1.4879838097647357e-05,
|
| 576 |
+
"loss": 3.2321,
|
| 577 |
+
"step": 2025
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.2592967366557045,
|
| 581 |
+
"grad_norm": 1.5469592809677124,
|
| 582 |
+
"learning_rate": 1.4816594991145967e-05,
|
| 583 |
+
"loss": 3.1697,
|
| 584 |
+
"step": 2050
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.2624588919807741,
|
| 588 |
+
"grad_norm": 1.7456170320510864,
|
| 589 |
+
"learning_rate": 1.4753351884644576e-05,
|
| 590 |
+
"loss": 3.0305,
|
| 591 |
+
"step": 2075
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.2656210473058437,
|
| 595 |
+
"grad_norm": 1.8613225221633911,
|
| 596 |
+
"learning_rate": 1.4690108778143184e-05,
|
| 597 |
+
"loss": 3.1094,
|
| 598 |
+
"step": 2100
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.2687832026309132,
|
| 602 |
+
"grad_norm": 1.6130377054214478,
|
| 603 |
+
"learning_rate": 1.4626865671641792e-05,
|
| 604 |
+
"loss": 3.1656,
|
| 605 |
+
"step": 2125
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.2719453579559828,
|
| 609 |
+
"grad_norm": 1.4752004146575928,
|
| 610 |
+
"learning_rate": 1.45636225651404e-05,
|
| 611 |
+
"loss": 3.1527,
|
| 612 |
+
"step": 2150
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.2751075132810524,
|
| 616 |
+
"grad_norm": 1.3546524047851562,
|
| 617 |
+
"learning_rate": 1.4500379458639008e-05,
|
| 618 |
+
"loss": 3.1096,
|
| 619 |
+
"step": 2175
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.27826966860612196,
|
| 623 |
+
"grad_norm": 1.5383223295211792,
|
| 624 |
+
"learning_rate": 1.4437136352137617e-05,
|
| 625 |
+
"loss": 3.3534,
|
| 626 |
+
"step": 2200
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.2814318239311915,
|
| 630 |
+
"grad_norm": 1.7905950546264648,
|
| 631 |
+
"learning_rate": 1.4373893245636228e-05,
|
| 632 |
+
"loss": 3.2779,
|
| 633 |
+
"step": 2225
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.28459397925626106,
|
| 637 |
+
"grad_norm": 1.408922553062439,
|
| 638 |
+
"learning_rate": 1.4310650139134836e-05,
|
| 639 |
+
"loss": 3.0517,
|
| 640 |
+
"step": 2250
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.28775613458133065,
|
| 644 |
+
"grad_norm": 1.7881851196289062,
|
| 645 |
+
"learning_rate": 1.4247407032633445e-05,
|
| 646 |
+
"loss": 3.144,
|
| 647 |
+
"step": 2275
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.29091828990640023,
|
| 651 |
+
"grad_norm": 1.5622018575668335,
|
| 652 |
+
"learning_rate": 1.4184163926132053e-05,
|
| 653 |
+
"loss": 3.0161,
|
| 654 |
+
"step": 2300
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.29408044523146976,
|
| 658 |
+
"grad_norm": 1.1805533170700073,
|
| 659 |
+
"learning_rate": 1.4120920819630661e-05,
|
| 660 |
+
"loss": 3.3535,
|
| 661 |
+
"step": 2325
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.29724260055653934,
|
| 665 |
+
"grad_norm": 1.4643152952194214,
|
| 666 |
+
"learning_rate": 1.4057677713129269e-05,
|
| 667 |
+
"loss": 3.0817,
|
| 668 |
+
"step": 2350
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.3004047558816089,
|
| 672 |
+
"grad_norm": 1.640499234199524,
|
| 673 |
+
"learning_rate": 1.3994434606627877e-05,
|
| 674 |
+
"loss": 3.2889,
|
| 675 |
+
"step": 2375
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.30356691120667845,
|
| 679 |
+
"grad_norm": 1.4501370191574097,
|
| 680 |
+
"learning_rate": 1.3931191500126486e-05,
|
| 681 |
+
"loss": 3.0626,
|
| 682 |
+
"step": 2400
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.30672906653174803,
|
| 686 |
+
"grad_norm": 1.540034532546997,
|
| 687 |
+
"learning_rate": 1.3867948393625097e-05,
|
| 688 |
+
"loss": 3.1131,
|
| 689 |
+
"step": 2425
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.3098912218568176,
|
| 693 |
+
"grad_norm": 1.3436205387115479,
|
| 694 |
+
"learning_rate": 1.3804705287123705e-05,
|
| 695 |
+
"loss": 3.3248,
|
| 696 |
+
"step": 2450
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.3130533771818872,
|
| 700 |
+
"grad_norm": 1.8110958337783813,
|
| 701 |
+
"learning_rate": 1.3741462180622314e-05,
|
| 702 |
+
"loss": 3.2063,
|
| 703 |
+
"step": 2475
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.3162155325069567,
|
| 707 |
+
"grad_norm": 1.1593376398086548,
|
| 708 |
+
"learning_rate": 1.3678219074120922e-05,
|
| 709 |
+
"loss": 3.1707,
|
| 710 |
+
"step": 2500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.3193776878320263,
|
| 714 |
+
"grad_norm": 1.2828359603881836,
|
| 715 |
+
"learning_rate": 1.361497596761953e-05,
|
| 716 |
+
"loss": 3.1273,
|
| 717 |
+
"step": 2525
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.3225398431570959,
|
| 721 |
+
"grad_norm": 1.4801135063171387,
|
| 722 |
+
"learning_rate": 1.3551732861118138e-05,
|
| 723 |
+
"loss": 3.1145,
|
| 724 |
+
"step": 2550
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.32570199848216547,
|
| 728 |
+
"grad_norm": 1.6385855674743652,
|
| 729 |
+
"learning_rate": 1.3488489754616746e-05,
|
| 730 |
+
"loss": 3.1274,
|
| 731 |
+
"step": 2575
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.328864153807235,
|
| 735 |
+
"grad_norm": 1.3759686946868896,
|
| 736 |
+
"learning_rate": 1.3425246648115358e-05,
|
| 737 |
+
"loss": 2.9515,
|
| 738 |
+
"step": 2600
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.3320263091323046,
|
| 742 |
+
"grad_norm": 1.3556058406829834,
|
| 743 |
+
"learning_rate": 1.3362003541613966e-05,
|
| 744 |
+
"loss": 3.1251,
|
| 745 |
+
"step": 2625
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.33518846445737416,
|
| 749 |
+
"grad_norm": 1.8028017282485962,
|
| 750 |
+
"learning_rate": 1.3298760435112574e-05,
|
| 751 |
+
"loss": 3.1851,
|
| 752 |
+
"step": 2650
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.33835061978244374,
|
| 756 |
+
"grad_norm": 1.3875707387924194,
|
| 757 |
+
"learning_rate": 1.3235517328611182e-05,
|
| 758 |
+
"loss": 3.1831,
|
| 759 |
+
"step": 2675
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.34151277510751327,
|
| 763 |
+
"grad_norm": 1.3739882707595825,
|
| 764 |
+
"learning_rate": 1.317227422210979e-05,
|
| 765 |
+
"loss": 2.944,
|
| 766 |
+
"step": 2700
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.34467493043258285,
|
| 770 |
+
"grad_norm": 1.7388805150985718,
|
| 771 |
+
"learning_rate": 1.3109031115608399e-05,
|
| 772 |
+
"loss": 2.9621,
|
| 773 |
+
"step": 2725
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.34783708575765243,
|
| 777 |
+
"grad_norm": 1.4218101501464844,
|
| 778 |
+
"learning_rate": 1.3045788009107007e-05,
|
| 779 |
+
"loss": 3.1734,
|
| 780 |
+
"step": 2750
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.35099924108272196,
|
| 784 |
+
"grad_norm": 1.5523903369903564,
|
| 785 |
+
"learning_rate": 1.2982544902605615e-05,
|
| 786 |
+
"loss": 3.1323,
|
| 787 |
+
"step": 2775
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.35416139640779154,
|
| 791 |
+
"grad_norm": 1.2134604454040527,
|
| 792 |
+
"learning_rate": 1.2919301796104227e-05,
|
| 793 |
+
"loss": 2.9738,
|
| 794 |
+
"step": 2800
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.3573235517328611,
|
| 798 |
+
"grad_norm": 1.7392768859863281,
|
| 799 |
+
"learning_rate": 1.2856058689602835e-05,
|
| 800 |
+
"loss": 3.2412,
|
| 801 |
+
"step": 2825
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.3604857070579307,
|
| 805 |
+
"grad_norm": 1.5656300783157349,
|
| 806 |
+
"learning_rate": 1.2792815583101443e-05,
|
| 807 |
+
"loss": 3.1549,
|
| 808 |
+
"step": 2850
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.36364786238300023,
|
| 812 |
+
"grad_norm": 1.592602014541626,
|
| 813 |
+
"learning_rate": 1.2729572476600051e-05,
|
| 814 |
+
"loss": 3.2068,
|
| 815 |
+
"step": 2875
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.3668100177080698,
|
| 819 |
+
"grad_norm": 1.6737302541732788,
|
| 820 |
+
"learning_rate": 1.266632937009866e-05,
|
| 821 |
+
"loss": 3.0918,
|
| 822 |
+
"step": 2900
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.3699721730331394,
|
| 826 |
+
"grad_norm": 1.4579912424087524,
|
| 827 |
+
"learning_rate": 1.2603086263597268e-05,
|
| 828 |
+
"loss": 3.17,
|
| 829 |
+
"step": 2925
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.373134328358209,
|
| 833 |
+
"grad_norm": 1.9524872303009033,
|
| 834 |
+
"learning_rate": 1.2539843157095876e-05,
|
| 835 |
+
"loss": 3.0157,
|
| 836 |
+
"step": 2950
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.3762964836832785,
|
| 840 |
+
"grad_norm": 1.5913100242614746,
|
| 841 |
+
"learning_rate": 1.2476600050594488e-05,
|
| 842 |
+
"loss": 2.9347,
|
| 843 |
+
"step": 2975
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.3794586390083481,
|
| 847 |
+
"grad_norm": 1.5421273708343506,
|
| 848 |
+
"learning_rate": 1.2413356944093096e-05,
|
| 849 |
+
"loss": 3.0643,
|
| 850 |
+
"step": 3000
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.38262079433341767,
|
| 854 |
+
"grad_norm": 1.4578076601028442,
|
| 855 |
+
"learning_rate": 1.2350113837591704e-05,
|
| 856 |
+
"loss": 3.1577,
|
| 857 |
+
"step": 3025
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.38578294965848725,
|
| 861 |
+
"grad_norm": 1.4034929275512695,
|
| 862 |
+
"learning_rate": 1.2286870731090312e-05,
|
| 863 |
+
"loss": 3.0522,
|
| 864 |
+
"step": 3050
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.3889451049835568,
|
| 868 |
+
"grad_norm": 1.4334189891815186,
|
| 869 |
+
"learning_rate": 1.222362762458892e-05,
|
| 870 |
+
"loss": 3.0195,
|
| 871 |
+
"step": 3075
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.39210726030862636,
|
| 875 |
+
"grad_norm": 1.5584102869033813,
|
| 876 |
+
"learning_rate": 1.2160384518087529e-05,
|
| 877 |
+
"loss": 2.9582,
|
| 878 |
+
"step": 3100
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.39526941563369594,
|
| 882 |
+
"grad_norm": 1.4117885828018188,
|
| 883 |
+
"learning_rate": 1.2097141411586137e-05,
|
| 884 |
+
"loss": 3.1979,
|
| 885 |
+
"step": 3125
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.39843157095876547,
|
| 889 |
+
"grad_norm": 1.6135624647140503,
|
| 890 |
+
"learning_rate": 1.2033898305084745e-05,
|
| 891 |
+
"loss": 3.0185,
|
| 892 |
+
"step": 3150
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.40159372628383505,
|
| 896 |
+
"grad_norm": 1.9495983123779297,
|
| 897 |
+
"learning_rate": 1.1970655198583357e-05,
|
| 898 |
+
"loss": 3.1753,
|
| 899 |
+
"step": 3175
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.40475588160890463,
|
| 903 |
+
"grad_norm": 1.3732932806015015,
|
| 904 |
+
"learning_rate": 1.1907412092081965e-05,
|
| 905 |
+
"loss": 2.9982,
|
| 906 |
+
"step": 3200
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.4079180369339742,
|
| 910 |
+
"grad_norm": 1.6417973041534424,
|
| 911 |
+
"learning_rate": 1.1844168985580573e-05,
|
| 912 |
+
"loss": 2.9491,
|
| 913 |
+
"step": 3225
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.41108019225904374,
|
| 917 |
+
"grad_norm": 1.599778413772583,
|
| 918 |
+
"learning_rate": 1.1780925879079181e-05,
|
| 919 |
+
"loss": 2.9619,
|
| 920 |
+
"step": 3250
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.4142423475841133,
|
| 924 |
+
"grad_norm": 1.5678138732910156,
|
| 925 |
+
"learning_rate": 1.171768277257779e-05,
|
| 926 |
+
"loss": 3.1179,
|
| 927 |
+
"step": 3275
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.4174045029091829,
|
| 931 |
+
"grad_norm": 1.4923375844955444,
|
| 932 |
+
"learning_rate": 1.1654439666076398e-05,
|
| 933 |
+
"loss": 3.193,
|
| 934 |
+
"step": 3300
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.4205666582342525,
|
| 938 |
+
"grad_norm": 1.5655521154403687,
|
| 939 |
+
"learning_rate": 1.1591196559575006e-05,
|
| 940 |
+
"loss": 3.0813,
|
| 941 |
+
"step": 3325
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.423728813559322,
|
| 945 |
+
"grad_norm": 1.7306265830993652,
|
| 946 |
+
"learning_rate": 1.1527953453073617e-05,
|
| 947 |
+
"loss": 3.1584,
|
| 948 |
+
"step": 3350
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.4268909688843916,
|
| 952 |
+
"grad_norm": 1.4964895248413086,
|
| 953 |
+
"learning_rate": 1.1464710346572226e-05,
|
| 954 |
+
"loss": 2.9885,
|
| 955 |
+
"step": 3375
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.4300531242094612,
|
| 959 |
+
"grad_norm": 1.386781930923462,
|
| 960 |
+
"learning_rate": 1.1401467240070834e-05,
|
| 961 |
+
"loss": 2.7611,
|
| 962 |
+
"step": 3400
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.43321527953453076,
|
| 966 |
+
"grad_norm": 1.223809838294983,
|
| 967 |
+
"learning_rate": 1.1338224133569442e-05,
|
| 968 |
+
"loss": 2.9793,
|
| 969 |
+
"step": 3425
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.4363774348596003,
|
| 973 |
+
"grad_norm": 1.7949477434158325,
|
| 974 |
+
"learning_rate": 1.127498102706805e-05,
|
| 975 |
+
"loss": 3.1053,
|
| 976 |
+
"step": 3450
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.43953959018466987,
|
| 980 |
+
"grad_norm": 1.2614104747772217,
|
| 981 |
+
"learning_rate": 1.1211737920566658e-05,
|
| 982 |
+
"loss": 3.1497,
|
| 983 |
+
"step": 3475
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.44270174550973945,
|
| 987 |
+
"grad_norm": 1.4934375286102295,
|
| 988 |
+
"learning_rate": 1.1148494814065267e-05,
|
| 989 |
+
"loss": 3.2349,
|
| 990 |
+
"step": 3500
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.445863900834809,
|
| 994 |
+
"grad_norm": 1.692384958267212,
|
| 995 |
+
"learning_rate": 1.1085251707563875e-05,
|
| 996 |
+
"loss": 3.2143,
|
| 997 |
+
"step": 3525
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.44902605615987856,
|
| 1001 |
+
"grad_norm": 1.430724024772644,
|
| 1002 |
+
"learning_rate": 1.1022008601062486e-05,
|
| 1003 |
+
"loss": 2.9926,
|
| 1004 |
+
"step": 3550
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.45218821148494814,
|
| 1008 |
+
"grad_norm": 1.6542253494262695,
|
| 1009 |
+
"learning_rate": 1.0958765494561095e-05,
|
| 1010 |
+
"loss": 3.0877,
|
| 1011 |
+
"step": 3575
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.4553503668100177,
|
| 1015 |
+
"grad_norm": 1.761777400970459,
|
| 1016 |
+
"learning_rate": 1.0895522388059703e-05,
|
| 1017 |
+
"loss": 3.1343,
|
| 1018 |
+
"step": 3600
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.45851252213508725,
|
| 1022 |
+
"grad_norm": 1.3432550430297852,
|
| 1023 |
+
"learning_rate": 1.0832279281558311e-05,
|
| 1024 |
+
"loss": 2.9125,
|
| 1025 |
+
"step": 3625
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.46167467746015683,
|
| 1029 |
+
"grad_norm": 1.6349095106124878,
|
| 1030 |
+
"learning_rate": 1.0769036175056919e-05,
|
| 1031 |
+
"loss": 3.0102,
|
| 1032 |
+
"step": 3650
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.4648368327852264,
|
| 1036 |
+
"grad_norm": 1.5871011018753052,
|
| 1037 |
+
"learning_rate": 1.0705793068555527e-05,
|
| 1038 |
+
"loss": 2.8361,
|
| 1039 |
+
"step": 3675
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.467998988110296,
|
| 1043 |
+
"grad_norm": 1.5581876039505005,
|
| 1044 |
+
"learning_rate": 1.0642549962054136e-05,
|
| 1045 |
+
"loss": 2.9176,
|
| 1046 |
+
"step": 3700
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.4711611434353655,
|
| 1050 |
+
"grad_norm": 1.9018642902374268,
|
| 1051 |
+
"learning_rate": 1.0579306855552747e-05,
|
| 1052 |
+
"loss": 3.1236,
|
| 1053 |
+
"step": 3725
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.4743232987604351,
|
| 1057 |
+
"grad_norm": 1.9724278450012207,
|
| 1058 |
+
"learning_rate": 1.0516063749051355e-05,
|
| 1059 |
+
"loss": 3.2517,
|
| 1060 |
+
"step": 3750
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.4774854540855047,
|
| 1064 |
+
"grad_norm": 1.4184868335723877,
|
| 1065 |
+
"learning_rate": 1.0452820642549963e-05,
|
| 1066 |
+
"loss": 2.7384,
|
| 1067 |
+
"step": 3775
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.4806476094105743,
|
| 1071 |
+
"grad_norm": 1.6251132488250732,
|
| 1072 |
+
"learning_rate": 1.0389577536048572e-05,
|
| 1073 |
+
"loss": 2.9232,
|
| 1074 |
+
"step": 3800
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.4838097647356438,
|
| 1078 |
+
"grad_norm": 1.526455283164978,
|
| 1079 |
+
"learning_rate": 1.032633442954718e-05,
|
| 1080 |
+
"loss": 3.0109,
|
| 1081 |
+
"step": 3825
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.4869719200607134,
|
| 1085 |
+
"grad_norm": 1.5482044219970703,
|
| 1086 |
+
"learning_rate": 1.0263091323045788e-05,
|
| 1087 |
+
"loss": 2.9056,
|
| 1088 |
+
"step": 3850
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.49013407538578296,
|
| 1092 |
+
"grad_norm": 1.5261505842208862,
|
| 1093 |
+
"learning_rate": 1.0199848216544396e-05,
|
| 1094 |
+
"loss": 2.8749,
|
| 1095 |
+
"step": 3875
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.4932962307108525,
|
| 1099 |
+
"grad_norm": 1.8025535345077515,
|
| 1100 |
+
"learning_rate": 1.0136605110043004e-05,
|
| 1101 |
+
"loss": 3.1128,
|
| 1102 |
+
"step": 3900
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.4964583860359221,
|
| 1106 |
+
"grad_norm": 1.428043246269226,
|
| 1107 |
+
"learning_rate": 1.0073362003541616e-05,
|
| 1108 |
+
"loss": 3.0411,
|
| 1109 |
+
"step": 3925
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.49962054136099165,
|
| 1113 |
+
"grad_norm": 1.3403562307357788,
|
| 1114 |
+
"learning_rate": 1.0010118897040224e-05,
|
| 1115 |
+
"loss": 2.9253,
|
| 1116 |
+
"step": 3950
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.5027826966860612,
|
| 1120 |
+
"grad_norm": 1.6576727628707886,
|
| 1121 |
+
"learning_rate": 9.946875790538832e-06,
|
| 1122 |
+
"loss": 2.9877,
|
| 1123 |
+
"step": 3975
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.5059448520111308,
|
| 1127 |
+
"grad_norm": 1.8511061668395996,
|
| 1128 |
+
"learning_rate": 9.88363268403744e-06,
|
| 1129 |
+
"loss": 2.919,
|
| 1130 |
+
"step": 4000
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.5091070073362004,
|
| 1134 |
+
"grad_norm": 1.690691590309143,
|
| 1135 |
+
"learning_rate": 9.820389577536049e-06,
|
| 1136 |
+
"loss": 2.7579,
|
| 1137 |
+
"step": 4025
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.5122691626612699,
|
| 1141 |
+
"grad_norm": 1.4383487701416016,
|
| 1142 |
+
"learning_rate": 9.757146471034657e-06,
|
| 1143 |
+
"loss": 2.8421,
|
| 1144 |
+
"step": 4050
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.5154313179863395,
|
| 1148 |
+
"grad_norm": 1.5412685871124268,
|
| 1149 |
+
"learning_rate": 9.693903364533267e-06,
|
| 1150 |
+
"loss": 2.9609,
|
| 1151 |
+
"step": 4075
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.518593473311409,
|
| 1155 |
+
"grad_norm": 1.7865098714828491,
|
| 1156 |
+
"learning_rate": 9.630660258031875e-06,
|
| 1157 |
+
"loss": 2.887,
|
| 1158 |
+
"step": 4100
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.5217556286364786,
|
| 1162 |
+
"grad_norm": 1.4453928470611572,
|
| 1163 |
+
"learning_rate": 9.567417151530483e-06,
|
| 1164 |
+
"loss": 2.9248,
|
| 1165 |
+
"step": 4125
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.5249177839615482,
|
| 1169 |
+
"grad_norm": 1.6136345863342285,
|
| 1170 |
+
"learning_rate": 9.504174045029093e-06,
|
| 1171 |
+
"loss": 3.1814,
|
| 1172 |
+
"step": 4150
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.5280799392866178,
|
| 1176 |
+
"grad_norm": 1.3460476398468018,
|
| 1177 |
+
"learning_rate": 9.440930938527701e-06,
|
| 1178 |
+
"loss": 2.7766,
|
| 1179 |
+
"step": 4175
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.5312420946116874,
|
| 1183 |
+
"grad_norm": 1.6284033060073853,
|
| 1184 |
+
"learning_rate": 9.37768783202631e-06,
|
| 1185 |
+
"loss": 2.9087,
|
| 1186 |
+
"step": 4200
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.534404249936757,
|
| 1190 |
+
"grad_norm": 1.3353221416473389,
|
| 1191 |
+
"learning_rate": 9.314444725524918e-06,
|
| 1192 |
+
"loss": 2.9512,
|
| 1193 |
+
"step": 4225
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.5375664052618264,
|
| 1197 |
+
"grad_norm": 1.9415779113769531,
|
| 1198 |
+
"learning_rate": 9.251201619023528e-06,
|
| 1199 |
+
"loss": 2.7603,
|
| 1200 |
+
"step": 4250
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.540728560586896,
|
| 1204 |
+
"grad_norm": 1.6942236423492432,
|
| 1205 |
+
"learning_rate": 9.187958512522136e-06,
|
| 1206 |
+
"loss": 3.1245,
|
| 1207 |
+
"step": 4275
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.5438907159119656,
|
| 1211 |
+
"grad_norm": 2.2894318103790283,
|
| 1212 |
+
"learning_rate": 9.124715406020744e-06,
|
| 1213 |
+
"loss": 2.9695,
|
| 1214 |
+
"step": 4300
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.5470528712370352,
|
| 1218 |
+
"grad_norm": 1.1514796018600464,
|
| 1219 |
+
"learning_rate": 9.061472299519352e-06,
|
| 1220 |
+
"loss": 3.0047,
|
| 1221 |
+
"step": 4325
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.5502150265621047,
|
| 1225 |
+
"grad_norm": 1.5051628351211548,
|
| 1226 |
+
"learning_rate": 8.998229193017962e-06,
|
| 1227 |
+
"loss": 3.1653,
|
| 1228 |
+
"step": 4350
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.5533771818871743,
|
| 1232 |
+
"grad_norm": 1.4090489149093628,
|
| 1233 |
+
"learning_rate": 8.93498608651657e-06,
|
| 1234 |
+
"loss": 2.9528,
|
| 1235 |
+
"step": 4375
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.5565393372122439,
|
| 1239 |
+
"grad_norm": 1.6863518953323364,
|
| 1240 |
+
"learning_rate": 8.871742980015179e-06,
|
| 1241 |
+
"loss": 2.8309,
|
| 1242 |
+
"step": 4400
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.5597014925373134,
|
| 1246 |
+
"grad_norm": 1.453518271446228,
|
| 1247 |
+
"learning_rate": 8.808499873513787e-06,
|
| 1248 |
+
"loss": 2.8803,
|
| 1249 |
+
"step": 4425
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.562863647862383,
|
| 1253 |
+
"grad_norm": 1.987695336341858,
|
| 1254 |
+
"learning_rate": 8.745256767012397e-06,
|
| 1255 |
+
"loss": 3.001,
|
| 1256 |
+
"step": 4450
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.5660258031874525,
|
| 1260 |
+
"grad_norm": 1.3925899267196655,
|
| 1261 |
+
"learning_rate": 8.682013660511005e-06,
|
| 1262 |
+
"loss": 2.9407,
|
| 1263 |
+
"step": 4475
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.5691879585125221,
|
| 1267 |
+
"grad_norm": 1.802440881729126,
|
| 1268 |
+
"learning_rate": 8.618770554009613e-06,
|
| 1269 |
+
"loss": 3.0096,
|
| 1270 |
+
"step": 4500
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.5723501138375917,
|
| 1274 |
+
"grad_norm": 1.4417994022369385,
|
| 1275 |
+
"learning_rate": 8.555527447508223e-06,
|
| 1276 |
+
"loss": 2.976,
|
| 1277 |
+
"step": 4525
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.5755122691626613,
|
| 1281 |
+
"grad_norm": 1.7502068281173706,
|
| 1282 |
+
"learning_rate": 8.492284341006831e-06,
|
| 1283 |
+
"loss": 2.8715,
|
| 1284 |
+
"step": 4550
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.5786744244877309,
|
| 1288 |
+
"grad_norm": 1.7311025857925415,
|
| 1289 |
+
"learning_rate": 8.42904123450544e-06,
|
| 1290 |
+
"loss": 2.9176,
|
| 1291 |
+
"step": 4575
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.5818365798128005,
|
| 1295 |
+
"grad_norm": 1.6848989725112915,
|
| 1296 |
+
"learning_rate": 8.365798128004048e-06,
|
| 1297 |
+
"loss": 2.9213,
|
| 1298 |
+
"step": 4600
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.5849987351378699,
|
| 1302 |
+
"grad_norm": 1.4339770078659058,
|
| 1303 |
+
"learning_rate": 8.302555021502657e-06,
|
| 1304 |
+
"loss": 2.944,
|
| 1305 |
+
"step": 4625
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.5881608904629395,
|
| 1309 |
+
"grad_norm": 1.4645488262176514,
|
| 1310 |
+
"learning_rate": 8.239311915001266e-06,
|
| 1311 |
+
"loss": 2.8865,
|
| 1312 |
+
"step": 4650
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.5913230457880091,
|
| 1316 |
+
"grad_norm": 1.793530821800232,
|
| 1317 |
+
"learning_rate": 8.176068808499874e-06,
|
| 1318 |
+
"loss": 3.0749,
|
| 1319 |
+
"step": 4675
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.5944852011130787,
|
| 1323 |
+
"grad_norm": 1.753974437713623,
|
| 1324 |
+
"learning_rate": 8.112825701998482e-06,
|
| 1325 |
+
"loss": 2.9768,
|
| 1326 |
+
"step": 4700
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.5976473564381483,
|
| 1330 |
+
"grad_norm": 1.4546597003936768,
|
| 1331 |
+
"learning_rate": 8.049582595497092e-06,
|
| 1332 |
+
"loss": 2.8419,
|
| 1333 |
+
"step": 4725
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.6008095117632178,
|
| 1337 |
+
"grad_norm": 2.1190099716186523,
|
| 1338 |
+
"learning_rate": 7.9863394889957e-06,
|
| 1339 |
+
"loss": 2.8468,
|
| 1340 |
+
"step": 4750
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.6039716670882874,
|
| 1344 |
+
"grad_norm": 2.026573657989502,
|
| 1345 |
+
"learning_rate": 7.923096382494308e-06,
|
| 1346 |
+
"loss": 3.0328,
|
| 1347 |
+
"step": 4775
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.6071338224133569,
|
| 1351 |
+
"grad_norm": 1.5188226699829102,
|
| 1352 |
+
"learning_rate": 7.859853275992917e-06,
|
| 1353 |
+
"loss": 2.7536,
|
| 1354 |
+
"step": 4800
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.6102959777384265,
|
| 1358 |
+
"grad_norm": 1.9499627351760864,
|
| 1359 |
+
"learning_rate": 7.796610169491526e-06,
|
| 1360 |
+
"loss": 3.0033,
|
| 1361 |
+
"step": 4825
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.6134581330634961,
|
| 1365 |
+
"grad_norm": 1.4692327976226807,
|
| 1366 |
+
"learning_rate": 7.733367062990135e-06,
|
| 1367 |
+
"loss": 2.9335,
|
| 1368 |
+
"step": 4850
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.6166202883885656,
|
| 1372 |
+
"grad_norm": 1.4884922504425049,
|
| 1373 |
+
"learning_rate": 7.670123956488743e-06,
|
| 1374 |
+
"loss": 2.9247,
|
| 1375 |
+
"step": 4875
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.6197824437136352,
|
| 1379 |
+
"grad_norm": 1.5420995950698853,
|
| 1380 |
+
"learning_rate": 7.606880849987353e-06,
|
| 1381 |
+
"loss": 2.9574,
|
| 1382 |
+
"step": 4900
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.6229445990387048,
|
| 1386 |
+
"grad_norm": 1.8658238649368286,
|
| 1387 |
+
"learning_rate": 7.543637743485961e-06,
|
| 1388 |
+
"loss": 2.9001,
|
| 1389 |
+
"step": 4925
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.6261067543637744,
|
| 1393 |
+
"grad_norm": 1.5181258916854858,
|
| 1394 |
+
"learning_rate": 7.480394636984569e-06,
|
| 1395 |
+
"loss": 2.7142,
|
| 1396 |
+
"step": 4950
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.629268909688844,
|
| 1400 |
+
"grad_norm": 1.4289120435714722,
|
| 1401 |
+
"learning_rate": 7.417151530483177e-06,
|
| 1402 |
+
"loss": 3.0176,
|
| 1403 |
+
"step": 4975
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.6324310650139134,
|
| 1407 |
+
"grad_norm": 1.7044678926467896,
|
| 1408 |
+
"learning_rate": 7.353908423981787e-06,
|
| 1409 |
+
"loss": 2.9688,
|
| 1410 |
+
"step": 5000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.635593220338983,
|
| 1414 |
+
"grad_norm": 1.2686058282852173,
|
| 1415 |
+
"learning_rate": 7.290665317480395e-06,
|
| 1416 |
+
"loss": 3.0132,
|
| 1417 |
+
"step": 5025
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.6387553756640526,
|
| 1421 |
+
"grad_norm": 1.4413508176803589,
|
| 1422 |
+
"learning_rate": 7.2274222109790036e-06,
|
| 1423 |
+
"loss": 2.948,
|
| 1424 |
+
"step": 5050
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.6419175309891222,
|
| 1428 |
+
"grad_norm": 1.5472356081008911,
|
| 1429 |
+
"learning_rate": 7.164179104477612e-06,
|
| 1430 |
+
"loss": 2.8153,
|
| 1431 |
+
"step": 5075
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.6450796863141918,
|
| 1435 |
+
"grad_norm": 1.3920260667800903,
|
| 1436 |
+
"learning_rate": 7.100935997976222e-06,
|
| 1437 |
+
"loss": 2.9269,
|
| 1438 |
+
"step": 5100
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.6482418416392614,
|
| 1442 |
+
"grad_norm": 1.8862273693084717,
|
| 1443 |
+
"learning_rate": 7.03769289147483e-06,
|
| 1444 |
+
"loss": 2.9641,
|
| 1445 |
+
"step": 5125
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.6514039969643309,
|
| 1449 |
+
"grad_norm": 1.2860732078552246,
|
| 1450 |
+
"learning_rate": 6.974449784973438e-06,
|
| 1451 |
+
"loss": 2.9628,
|
| 1452 |
+
"step": 5150
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.6545661522894004,
|
| 1456 |
+
"grad_norm": 1.6975411176681519,
|
| 1457 |
+
"learning_rate": 6.911206678472046e-06,
|
| 1458 |
+
"loss": 2.8478,
|
| 1459 |
+
"step": 5175
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.65772830761447,
|
| 1463 |
+
"grad_norm": 1.577736258506775,
|
| 1464 |
+
"learning_rate": 6.847963571970656e-06,
|
| 1465 |
+
"loss": 2.8673,
|
| 1466 |
+
"step": 5200
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.6608904629395396,
|
| 1470 |
+
"grad_norm": 1.5640876293182373,
|
| 1471 |
+
"learning_rate": 6.784720465469264e-06,
|
| 1472 |
+
"loss": 2.7907,
|
| 1473 |
+
"step": 5225
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.6640526182646092,
|
| 1477 |
+
"grad_norm": 1.5373408794403076,
|
| 1478 |
+
"learning_rate": 6.7214773589678725e-06,
|
| 1479 |
+
"loss": 2.9308,
|
| 1480 |
+
"step": 5250
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.6672147735896787,
|
| 1484 |
+
"grad_norm": 1.499895453453064,
|
| 1485 |
+
"learning_rate": 6.6582342524664824e-06,
|
| 1486 |
+
"loss": 2.8954,
|
| 1487 |
+
"step": 5275
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.6703769289147483,
|
| 1491 |
+
"grad_norm": 1.7884222269058228,
|
| 1492 |
+
"learning_rate": 6.594991145965091e-06,
|
| 1493 |
+
"loss": 3.0374,
|
| 1494 |
+
"step": 5300
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.6735390842398179,
|
| 1498 |
+
"grad_norm": 1.4689120054244995,
|
| 1499 |
+
"learning_rate": 6.531748039463699e-06,
|
| 1500 |
+
"loss": 2.7937,
|
| 1501 |
+
"step": 5325
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.6767012395648875,
|
| 1505 |
+
"grad_norm": 1.2251571416854858,
|
| 1506 |
+
"learning_rate": 6.468504932962307e-06,
|
| 1507 |
+
"loss": 2.6993,
|
| 1508 |
+
"step": 5350
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.679863394889957,
|
| 1512 |
+
"grad_norm": 1.6900901794433594,
|
| 1513 |
+
"learning_rate": 6.405261826460917e-06,
|
| 1514 |
+
"loss": 2.9683,
|
| 1515 |
+
"step": 5375
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.6830255502150265,
|
| 1519 |
+
"grad_norm": 1.3138439655303955,
|
| 1520 |
+
"learning_rate": 6.342018719959525e-06,
|
| 1521 |
+
"loss": 2.9179,
|
| 1522 |
+
"step": 5400
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.6861877055400961,
|
| 1526 |
+
"grad_norm": 1.3106989860534668,
|
| 1527 |
+
"learning_rate": 6.278775613458133e-06,
|
| 1528 |
+
"loss": 2.8901,
|
| 1529 |
+
"step": 5425
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.6893498608651657,
|
| 1533 |
+
"grad_norm": 1.7244914770126343,
|
| 1534 |
+
"learning_rate": 6.2155325069567415e-06,
|
| 1535 |
+
"loss": 2.8481,
|
| 1536 |
+
"step": 5450
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.6925120161902353,
|
| 1540 |
+
"grad_norm": 1.1945232152938843,
|
| 1541 |
+
"learning_rate": 6.152289400455351e-06,
|
| 1542 |
+
"loss": 2.8619,
|
| 1543 |
+
"step": 5475
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.6956741715153049,
|
| 1547 |
+
"grad_norm": 1.3923369646072388,
|
| 1548 |
+
"learning_rate": 6.08904629395396e-06,
|
| 1549 |
+
"loss": 2.9955,
|
| 1550 |
+
"step": 5500
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.6988363268403744,
|
| 1554 |
+
"grad_norm": 1.7948927879333496,
|
| 1555 |
+
"learning_rate": 6.025803187452568e-06,
|
| 1556 |
+
"loss": 3.0742,
|
| 1557 |
+
"step": 5525
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.7019984821654439,
|
| 1561 |
+
"grad_norm": 1.4123344421386719,
|
| 1562 |
+
"learning_rate": 5.962560080951176e-06,
|
| 1563 |
+
"loss": 2.9499,
|
| 1564 |
+
"step": 5550
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.7051606374905135,
|
| 1568 |
+
"grad_norm": 1.5082272291183472,
|
| 1569 |
+
"learning_rate": 5.899316974449786e-06,
|
| 1570 |
+
"loss": 3.0861,
|
| 1571 |
+
"step": 5575
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.7083227928155831,
|
| 1575 |
+
"grad_norm": 1.6818978786468506,
|
| 1576 |
+
"learning_rate": 5.836073867948394e-06,
|
| 1577 |
+
"loss": 2.7944,
|
| 1578 |
+
"step": 5600
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.7114849481406527,
|
| 1582 |
+
"grad_norm": 2.072303295135498,
|
| 1583 |
+
"learning_rate": 5.772830761447002e-06,
|
| 1584 |
+
"loss": 3.021,
|
| 1585 |
+
"step": 5625
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.7146471034657222,
|
| 1589 |
+
"grad_norm": 1.2416434288024902,
|
| 1590 |
+
"learning_rate": 5.7095876549456105e-06,
|
| 1591 |
+
"loss": 2.7849,
|
| 1592 |
+
"step": 5650
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.7178092587907918,
|
| 1596 |
+
"grad_norm": 1.6787675619125366,
|
| 1597 |
+
"learning_rate": 5.64634454844422e-06,
|
| 1598 |
+
"loss": 2.9102,
|
| 1599 |
+
"step": 5675
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.7209714141158614,
|
| 1603 |
+
"grad_norm": 1.542455792427063,
|
| 1604 |
+
"learning_rate": 5.5831014419428286e-06,
|
| 1605 |
+
"loss": 2.8268,
|
| 1606 |
+
"step": 5700
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.724133569440931,
|
| 1610 |
+
"grad_norm": 2.0236566066741943,
|
| 1611 |
+
"learning_rate": 5.519858335441437e-06,
|
| 1612 |
+
"loss": 2.8222,
|
| 1613 |
+
"step": 5725
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.7272957247660005,
|
| 1617 |
+
"grad_norm": 1.5775691270828247,
|
| 1618 |
+
"learning_rate": 5.456615228940047e-06,
|
| 1619 |
+
"loss": 2.9778,
|
| 1620 |
+
"step": 5750
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.73045788009107,
|
| 1624 |
+
"grad_norm": 1.792724847793579,
|
| 1625 |
+
"learning_rate": 5.393372122438655e-06,
|
| 1626 |
+
"loss": 2.9323,
|
| 1627 |
+
"step": 5775
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.7336200354161396,
|
| 1631 |
+
"grad_norm": 1.5170587301254272,
|
| 1632 |
+
"learning_rate": 5.330129015937263e-06,
|
| 1633 |
+
"loss": 2.8671,
|
| 1634 |
+
"step": 5800
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.7367821907412092,
|
| 1638 |
+
"grad_norm": 1.418395757675171,
|
| 1639 |
+
"learning_rate": 5.266885909435871e-06,
|
| 1640 |
+
"loss": 2.8153,
|
| 1641 |
+
"step": 5825
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.7399443460662788,
|
| 1645 |
+
"grad_norm": 1.4458298683166504,
|
| 1646 |
+
"learning_rate": 5.203642802934481e-06,
|
| 1647 |
+
"loss": 2.9178,
|
| 1648 |
+
"step": 5850
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.7431065013913484,
|
| 1652 |
+
"grad_norm": 1.6139628887176514,
|
| 1653 |
+
"learning_rate": 5.140399696433089e-06,
|
| 1654 |
+
"loss": 2.8857,
|
| 1655 |
+
"step": 5875
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.746268656716418,
|
| 1659 |
+
"grad_norm": 1.1631035804748535,
|
| 1660 |
+
"learning_rate": 5.0771565899316975e-06,
|
| 1661 |
+
"loss": 2.9213,
|
| 1662 |
+
"step": 5900
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.7494308120414874,
|
| 1666 |
+
"grad_norm": 1.5703893899917603,
|
| 1667 |
+
"learning_rate": 5.013913483430306e-06,
|
| 1668 |
+
"loss": 2.891,
|
| 1669 |
+
"step": 5925
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.752592967366557,
|
| 1673 |
+
"grad_norm": 1.4471451044082642,
|
| 1674 |
+
"learning_rate": 4.950670376928915e-06,
|
| 1675 |
+
"loss": 2.8261,
|
| 1676 |
+
"step": 5950
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.7557551226916266,
|
| 1680 |
+
"grad_norm": 1.6305294036865234,
|
| 1681 |
+
"learning_rate": 4.887427270427524e-06,
|
| 1682 |
+
"loss": 2.8918,
|
| 1683 |
+
"step": 5975
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.7589172780166962,
|
| 1687 |
+
"grad_norm": 1.3186579942703247,
|
| 1688 |
+
"learning_rate": 4.824184163926132e-06,
|
| 1689 |
+
"loss": 2.9478,
|
| 1690 |
+
"step": 6000
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.7620794333417658,
|
| 1694 |
+
"grad_norm": 1.4258198738098145,
|
| 1695 |
+
"learning_rate": 4.760941057424741e-06,
|
| 1696 |
+
"loss": 2.9313,
|
| 1697 |
+
"step": 6025
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.7652415886668353,
|
| 1701 |
+
"grad_norm": 1.5036190748214722,
|
| 1702 |
+
"learning_rate": 4.69769795092335e-06,
|
| 1703 |
+
"loss": 2.9562,
|
| 1704 |
+
"step": 6050
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.7684037439919049,
|
| 1708 |
+
"grad_norm": 1.7764675617218018,
|
| 1709 |
+
"learning_rate": 4.634454844421958e-06,
|
| 1710 |
+
"loss": 2.9396,
|
| 1711 |
+
"step": 6075
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.7715658993169745,
|
| 1715 |
+
"grad_norm": 1.624964714050293,
|
| 1716 |
+
"learning_rate": 4.571211737920567e-06,
|
| 1717 |
+
"loss": 2.8789,
|
| 1718 |
+
"step": 6100
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.774728054642044,
|
| 1722 |
+
"grad_norm": 1.1885554790496826,
|
| 1723 |
+
"learning_rate": 4.5079686314191755e-06,
|
| 1724 |
+
"loss": 2.9756,
|
| 1725 |
+
"step": 6125
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.7778902099671136,
|
| 1729 |
+
"grad_norm": 1.752863883972168,
|
| 1730 |
+
"learning_rate": 4.444725524917785e-06,
|
| 1731 |
+
"loss": 2.7279,
|
| 1732 |
+
"step": 6150
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.7810523652921831,
|
| 1736 |
+
"grad_norm": 1.4653725624084473,
|
| 1737 |
+
"learning_rate": 4.381482418416393e-06,
|
| 1738 |
+
"loss": 3.0123,
|
| 1739 |
+
"step": 6175
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.7842145206172527,
|
| 1743 |
+
"grad_norm": 1.5357298851013184,
|
| 1744 |
+
"learning_rate": 4.318239311915002e-06,
|
| 1745 |
+
"loss": 2.7978,
|
| 1746 |
+
"step": 6200
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.7873766759423223,
|
| 1750 |
+
"grad_norm": 1.4677993059158325,
|
| 1751 |
+
"learning_rate": 4.25499620541361e-06,
|
| 1752 |
+
"loss": 2.8016,
|
| 1753 |
+
"step": 6225
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.7905388312673919,
|
| 1757 |
+
"grad_norm": 1.3810391426086426,
|
| 1758 |
+
"learning_rate": 4.191753098912219e-06,
|
| 1759 |
+
"loss": 2.7919,
|
| 1760 |
+
"step": 6250
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.7937009865924615,
|
| 1764 |
+
"grad_norm": 1.827034831047058,
|
| 1765 |
+
"learning_rate": 4.128509992410827e-06,
|
| 1766 |
+
"loss": 2.9552,
|
| 1767 |
+
"step": 6275
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.7968631419175309,
|
| 1771 |
+
"grad_norm": 1.7732640504837036,
|
| 1772 |
+
"learning_rate": 4.065266885909436e-06,
|
| 1773 |
+
"loss": 3.03,
|
| 1774 |
+
"step": 6300
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.8000252972426005,
|
| 1778 |
+
"grad_norm": 2.2995574474334717,
|
| 1779 |
+
"learning_rate": 4.0020237794080445e-06,
|
| 1780 |
+
"loss": 3.027,
|
| 1781 |
+
"step": 6325
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.8031874525676701,
|
| 1785 |
+
"grad_norm": 1.664284110069275,
|
| 1786 |
+
"learning_rate": 3.9387806729066536e-06,
|
| 1787 |
+
"loss": 2.9782,
|
| 1788 |
+
"step": 6350
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.8063496078927397,
|
| 1792 |
+
"grad_norm": 1.7057468891143799,
|
| 1793 |
+
"learning_rate": 3.875537566405262e-06,
|
| 1794 |
+
"loss": 2.8451,
|
| 1795 |
+
"step": 6375
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.8095117632178093,
|
| 1799 |
+
"grad_norm": 1.548420786857605,
|
| 1800 |
+
"learning_rate": 3.812294459903871e-06,
|
| 1801 |
+
"loss": 2.7688,
|
| 1802 |
+
"step": 6400
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.8126739185428788,
|
| 1806 |
+
"grad_norm": 1.467391014099121,
|
| 1807 |
+
"learning_rate": 3.749051353402479e-06,
|
| 1808 |
+
"loss": 2.9415,
|
| 1809 |
+
"step": 6425
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.8158360738679484,
|
| 1813 |
+
"grad_norm": 1.6299670934677124,
|
| 1814 |
+
"learning_rate": 3.685808246901088e-06,
|
| 1815 |
+
"loss": 2.9104,
|
| 1816 |
+
"step": 6450
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.818998229193018,
|
| 1820 |
+
"grad_norm": 1.7904433012008667,
|
| 1821 |
+
"learning_rate": 3.622565140399697e-06,
|
| 1822 |
+
"loss": 2.8673,
|
| 1823 |
+
"step": 6475
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.8221603845180875,
|
| 1827 |
+
"grad_norm": 1.5487452745437622,
|
| 1828 |
+
"learning_rate": 3.5593220338983053e-06,
|
| 1829 |
+
"loss": 2.991,
|
| 1830 |
+
"step": 6500
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.8253225398431571,
|
| 1834 |
+
"grad_norm": 2.021498918533325,
|
| 1835 |
+
"learning_rate": 3.4960789273969143e-06,
|
| 1836 |
+
"loss": 2.9309,
|
| 1837 |
+
"step": 6525
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.8284846951682266,
|
| 1841 |
+
"grad_norm": 1.6797212362289429,
|
| 1842 |
+
"learning_rate": 3.4328358208955225e-06,
|
| 1843 |
+
"loss": 2.7787,
|
| 1844 |
+
"step": 6550
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.8316468504932962,
|
| 1848 |
+
"grad_norm": 1.5939486026763916,
|
| 1849 |
+
"learning_rate": 3.3695927143941316e-06,
|
| 1850 |
+
"loss": 2.8101,
|
| 1851 |
+
"step": 6575
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.8348090058183658,
|
| 1855 |
+
"grad_norm": 1.506430983543396,
|
| 1856 |
+
"learning_rate": 3.3063496078927398e-06,
|
| 1857 |
+
"loss": 2.9788,
|
| 1858 |
+
"step": 6600
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.8379711611434354,
|
| 1862 |
+
"grad_norm": 1.3282697200775146,
|
| 1863 |
+
"learning_rate": 3.243106501391349e-06,
|
| 1864 |
+
"loss": 2.8371,
|
| 1865 |
+
"step": 6625
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.841133316468505,
|
| 1869 |
+
"grad_norm": 1.799822449684143,
|
| 1870 |
+
"learning_rate": 3.179863394889957e-06,
|
| 1871 |
+
"loss": 2.9774,
|
| 1872 |
+
"step": 6650
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.8442954717935744,
|
| 1876 |
+
"grad_norm": 1.6544405221939087,
|
| 1877 |
+
"learning_rate": 3.116620288388566e-06,
|
| 1878 |
+
"loss": 2.8366,
|
| 1879 |
+
"step": 6675
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.847457627118644,
|
| 1883 |
+
"grad_norm": 1.7522649765014648,
|
| 1884 |
+
"learning_rate": 3.0533771818871742e-06,
|
| 1885 |
+
"loss": 2.9712,
|
| 1886 |
+
"step": 6700
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.8506197824437136,
|
| 1890 |
+
"grad_norm": 1.5311869382858276,
|
| 1891 |
+
"learning_rate": 2.9901340753857833e-06,
|
| 1892 |
+
"loss": 2.792,
|
| 1893 |
+
"step": 6725
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.8537819377687832,
|
| 1897 |
+
"grad_norm": 1.1723778247833252,
|
| 1898 |
+
"learning_rate": 2.9268909688843915e-06,
|
| 1899 |
+
"loss": 2.9395,
|
| 1900 |
+
"step": 6750
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.8569440930938528,
|
| 1904 |
+
"grad_norm": 1.9257378578186035,
|
| 1905 |
+
"learning_rate": 2.8636478623830005e-06,
|
| 1906 |
+
"loss": 2.9973,
|
| 1907 |
+
"step": 6775
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.8601062484189224,
|
| 1911 |
+
"grad_norm": 1.4320708513259888,
|
| 1912 |
+
"learning_rate": 2.8004047558816087e-06,
|
| 1913 |
+
"loss": 2.739,
|
| 1914 |
+
"step": 6800
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.8632684037439919,
|
| 1918 |
+
"grad_norm": 1.961350679397583,
|
| 1919 |
+
"learning_rate": 2.7371616493802178e-06,
|
| 1920 |
+
"loss": 3.0129,
|
| 1921 |
+
"step": 6825
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.8664305590690615,
|
| 1925 |
+
"grad_norm": 1.7785652875900269,
|
| 1926 |
+
"learning_rate": 2.673918542878827e-06,
|
| 1927 |
+
"loss": 3.0007,
|
| 1928 |
+
"step": 6850
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.869592714394131,
|
| 1932 |
+
"grad_norm": 1.5985599756240845,
|
| 1933 |
+
"learning_rate": 2.610675436377435e-06,
|
| 1934 |
+
"loss": 2.8996,
|
| 1935 |
+
"step": 6875
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.8727548697192006,
|
| 1939 |
+
"grad_norm": 1.4986529350280762,
|
| 1940 |
+
"learning_rate": 2.547432329876044e-06,
|
| 1941 |
+
"loss": 2.917,
|
| 1942 |
+
"step": 6900
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.8759170250442702,
|
| 1946 |
+
"grad_norm": 1.4311537742614746,
|
| 1947 |
+
"learning_rate": 2.4841892233746523e-06,
|
| 1948 |
+
"loss": 2.8844,
|
| 1949 |
+
"step": 6925
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.8790791803693397,
|
| 1953 |
+
"grad_norm": 1.8189743757247925,
|
| 1954 |
+
"learning_rate": 2.420946116873261e-06,
|
| 1955 |
+
"loss": 2.9588,
|
| 1956 |
+
"step": 6950
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.8822413356944093,
|
| 1960 |
+
"grad_norm": 1.1215569972991943,
|
| 1961 |
+
"learning_rate": 2.3577030103718695e-06,
|
| 1962 |
+
"loss": 2.8752,
|
| 1963 |
+
"step": 6975
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.8854034910194789,
|
| 1967 |
+
"grad_norm": 1.4951814413070679,
|
| 1968 |
+
"learning_rate": 2.294459903870478e-06,
|
| 1969 |
+
"loss": 2.8954,
|
| 1970 |
+
"step": 7000
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.8885656463445485,
|
| 1974 |
+
"grad_norm": 1.464539885520935,
|
| 1975 |
+
"learning_rate": 2.231216797369087e-06,
|
| 1976 |
+
"loss": 2.7321,
|
| 1977 |
+
"step": 7025
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.891727801669618,
|
| 1981 |
+
"grad_norm": 1.4621193408966064,
|
| 1982 |
+
"learning_rate": 2.1679736908676958e-06,
|
| 1983 |
+
"loss": 2.8688,
|
| 1984 |
+
"step": 7050
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.8948899569946875,
|
| 1988 |
+
"grad_norm": 1.6291460990905762,
|
| 1989 |
+
"learning_rate": 2.1047305843663044e-06,
|
| 1990 |
+
"loss": 2.8068,
|
| 1991 |
+
"step": 7075
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.8980521123197571,
|
| 1995 |
+
"grad_norm": 1.4216442108154297,
|
| 1996 |
+
"learning_rate": 2.041487477864913e-06,
|
| 1997 |
+
"loss": 2.9351,
|
| 1998 |
+
"step": 7100
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 0.9012142676448267,
|
| 2002 |
+
"grad_norm": 1.306211233139038,
|
| 2003 |
+
"learning_rate": 1.9782443713635216e-06,
|
| 2004 |
+
"loss": 2.9398,
|
| 2005 |
+
"step": 7125
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.9043764229698963,
|
| 2009 |
+
"grad_norm": 1.6560629606246948,
|
| 2010 |
+
"learning_rate": 1.9150012648621303e-06,
|
| 2011 |
+
"loss": 2.9728,
|
| 2012 |
+
"step": 7150
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.9075385782949659,
|
| 2016 |
+
"grad_norm": 1.9747501611709595,
|
| 2017 |
+
"learning_rate": 1.8517581583607389e-06,
|
| 2018 |
+
"loss": 2.903,
|
| 2019 |
+
"step": 7175
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 0.9107007336200355,
|
| 2023 |
+
"grad_norm": 1.3126964569091797,
|
| 2024 |
+
"learning_rate": 1.7885150518593475e-06,
|
| 2025 |
+
"loss": 2.796,
|
| 2026 |
+
"step": 7200
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.913862888945105,
|
| 2030 |
+
"grad_norm": 1.5026005506515503,
|
| 2031 |
+
"learning_rate": 1.7252719453579561e-06,
|
| 2032 |
+
"loss": 2.7565,
|
| 2033 |
+
"step": 7225
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 0.9170250442701745,
|
| 2037 |
+
"grad_norm": 1.5117318630218506,
|
| 2038 |
+
"learning_rate": 1.6620288388565647e-06,
|
| 2039 |
+
"loss": 2.8329,
|
| 2040 |
+
"step": 7250
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 0.9201871995952441,
|
| 2044 |
+
"grad_norm": 1.812044382095337,
|
| 2045 |
+
"learning_rate": 1.5987857323551734e-06,
|
| 2046 |
+
"loss": 2.9988,
|
| 2047 |
+
"step": 7275
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 0.9233493549203137,
|
| 2051 |
+
"grad_norm": 1.25447678565979,
|
| 2052 |
+
"learning_rate": 1.535542625853782e-06,
|
| 2053 |
+
"loss": 2.9395,
|
| 2054 |
+
"step": 7300
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 0.9265115102453833,
|
| 2058 |
+
"grad_norm": 1.6009807586669922,
|
| 2059 |
+
"learning_rate": 1.4722995193523906e-06,
|
| 2060 |
+
"loss": 2.8487,
|
| 2061 |
+
"step": 7325
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 0.9296736655704528,
|
| 2065 |
+
"grad_norm": 2.0983338356018066,
|
| 2066 |
+
"learning_rate": 1.4090564128509992e-06,
|
| 2067 |
+
"loss": 2.841,
|
| 2068 |
+
"step": 7350
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 0.9328358208955224,
|
| 2072 |
+
"grad_norm": 1.5755832195281982,
|
| 2073 |
+
"learning_rate": 1.3458133063496079e-06,
|
| 2074 |
+
"loss": 2.8646,
|
| 2075 |
+
"step": 7375
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 0.935997976220592,
|
| 2079 |
+
"grad_norm": 1.4814996719360352,
|
| 2080 |
+
"learning_rate": 1.2825701998482165e-06,
|
| 2081 |
+
"loss": 3.0229,
|
| 2082 |
+
"step": 7400
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.9391601315456615,
|
| 2086 |
+
"grad_norm": 1.687947392463684,
|
| 2087 |
+
"learning_rate": 1.2193270933468253e-06,
|
| 2088 |
+
"loss": 2.8668,
|
| 2089 |
+
"step": 7425
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 0.942322286870731,
|
| 2093 |
+
"grad_norm": 1.557085394859314,
|
| 2094 |
+
"learning_rate": 1.156083986845434e-06,
|
| 2095 |
+
"loss": 2.9006,
|
| 2096 |
+
"step": 7450
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.9454844421958006,
|
| 2100 |
+
"grad_norm": 1.4846750497817993,
|
| 2101 |
+
"learning_rate": 1.0928408803440425e-06,
|
| 2102 |
+
"loss": 2.9055,
|
| 2103 |
+
"step": 7475
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 0.9486465975208702,
|
| 2107 |
+
"grad_norm": 2.095038652420044,
|
| 2108 |
+
"learning_rate": 1.0295977738426512e-06,
|
| 2109 |
+
"loss": 3.0127,
|
| 2110 |
+
"step": 7500
|
| 2111 |
+
}
|
| 2112 |
+
],
|
| 2113 |
+
"logging_steps": 25,
|
| 2114 |
+
"max_steps": 7906,
|
| 2115 |
+
"num_input_tokens_seen": 0,
|
| 2116 |
+
"num_train_epochs": 1,
|
| 2117 |
+
"save_steps": 500,
|
| 2118 |
+
"stateful_callbacks": {
|
| 2119 |
+
"TrainerControl": {
|
| 2120 |
+
"args": {
|
| 2121 |
+
"should_epoch_stop": false,
|
| 2122 |
+
"should_evaluate": false,
|
| 2123 |
+
"should_log": false,
|
| 2124 |
+
"should_save": true,
|
| 2125 |
+
"should_training_stop": false
|
| 2126 |
+
},
|
| 2127 |
+
"attributes": {}
|
| 2128 |
+
}
|
| 2129 |
+
},
|
| 2130 |
+
"total_flos": 1995574026240000.0,
|
| 2131 |
+
"train_batch_size": 2,
|
| 2132 |
+
"trial_name": null,
|
| 2133 |
+
"trial_params": null
|
| 2134 |
+
}
|
checkpoint-7500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:72b581cd8e50466800ca90569133129aafadaef26c6c1972c6abcf6f033d31ec
|
| 3 |
+
size 5869
|
checkpoint-7906/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_cross_attention": true,
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertLMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"gradient_checkpointing": false,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 384,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 1536,
|
| 15 |
+
"is_decoder": true,
|
| 16 |
+
"layer_norm_eps": 1e-12,
|
| 17 |
+
"max_position_embeddings": 512,
|
| 18 |
+
"model_type": "bert",
|
| 19 |
+
"num_attention_heads": 12,
|
| 20 |
+
"num_hidden_layers": 6,
|
| 21 |
+
"pad_token_id": 0,
|
| 22 |
+
"position_embedding_type": "absolute",
|
| 23 |
+
"transformers_version": "4.57.1",
|
| 24 |
+
"type_vocab_size": 2,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 30522
|
| 27 |
+
}
|
checkpoint-7906/generation_config.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"pad_token_id": 0,
|
| 4 |
+
"transformers_version": "4.57.1"
|
| 5 |
+
}
|
checkpoint-7906/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a16e5f60ed44d8542a97a281d474481a19da2f9891decde86a56e18e6ae81b9
|
| 3 |
+
size 90990200
|
checkpoint-7906/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dc6b9409750e7722b14804716a7cd8c38c9ce22af04b1d1a0261ebfef6391c3a
|
| 3 |
+
size 182045763
|
checkpoint-7906/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c4d99ee630a976bb8c7d6176de77b819582254d49513fc0d7b379a9da346929
|
| 3 |
+
size 14565
|
checkpoint-7906/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:05d93cfea447e9e85726e69ff11d8f88b5af6417bfc064d3e1d3d5151db8ea84
|
| 3 |
+
size 1389
|
checkpoint-7906/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b12c834264ad82dd87e7717e3f15cfb9e42b004d4cceeda0c654f0fa220db40e
|
| 3 |
+
size 1453
|
checkpoint-7906/trainer_state.json
ADDED
|
@@ -0,0 +1,2246 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 7906,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0031621553250695674,
|
| 14 |
+
"grad_norm": 11.55327320098877,
|
| 15 |
+
"learning_rate": 1.9939286617758668e-05,
|
| 16 |
+
"loss": 8.596,
|
| 17 |
+
"step": 25
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.006324310650139135,
|
| 21 |
+
"grad_norm": 5.928526401519775,
|
| 22 |
+
"learning_rate": 1.9876043511257274e-05,
|
| 23 |
+
"loss": 5.501,
|
| 24 |
+
"step": 50
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.009486465975208701,
|
| 28 |
+
"grad_norm": 2.7216451168060303,
|
| 29 |
+
"learning_rate": 1.9812800404755884e-05,
|
| 30 |
+
"loss": 4.9742,
|
| 31 |
+
"step": 75
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.01264862130027827,
|
| 35 |
+
"grad_norm": 2.725499153137207,
|
| 36 |
+
"learning_rate": 1.974955729825449e-05,
|
| 37 |
+
"loss": 4.6023,
|
| 38 |
+
"step": 100
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.015810776625347838,
|
| 42 |
+
"grad_norm": 2.318068027496338,
|
| 43 |
+
"learning_rate": 1.96863141917531e-05,
|
| 44 |
+
"loss": 4.4124,
|
| 45 |
+
"step": 125
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.018972931950417403,
|
| 49 |
+
"grad_norm": 2.437415838241577,
|
| 50 |
+
"learning_rate": 1.9623071085251707e-05,
|
| 51 |
+
"loss": 4.5471,
|
| 52 |
+
"step": 150
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.02213508727548697,
|
| 56 |
+
"grad_norm": 1.6126755475997925,
|
| 57 |
+
"learning_rate": 1.955982797875032e-05,
|
| 58 |
+
"loss": 4.0951,
|
| 59 |
+
"step": 175
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.02529724260055654,
|
| 63 |
+
"grad_norm": 1.9190938472747803,
|
| 64 |
+
"learning_rate": 1.9496584872248927e-05,
|
| 65 |
+
"loss": 4.0572,
|
| 66 |
+
"step": 200
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.028459397925626108,
|
| 70 |
+
"grad_norm": 1.6798152923583984,
|
| 71 |
+
"learning_rate": 1.9433341765747537e-05,
|
| 72 |
+
"loss": 3.9318,
|
| 73 |
+
"step": 225
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.031621553250695676,
|
| 77 |
+
"grad_norm": 2.02826189994812,
|
| 78 |
+
"learning_rate": 1.9370098659246143e-05,
|
| 79 |
+
"loss": 3.9254,
|
| 80 |
+
"step": 250
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.034783708575765244,
|
| 84 |
+
"grad_norm": 1.6234019994735718,
|
| 85 |
+
"learning_rate": 1.9306855552744753e-05,
|
| 86 |
+
"loss": 3.7144,
|
| 87 |
+
"step": 275
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.037945863900834806,
|
| 91 |
+
"grad_norm": 2.1222445964813232,
|
| 92 |
+
"learning_rate": 1.924361244624336e-05,
|
| 93 |
+
"loss": 4.0748,
|
| 94 |
+
"step": 300
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.041108019225904374,
|
| 98 |
+
"grad_norm": 1.8071123361587524,
|
| 99 |
+
"learning_rate": 1.918036933974197e-05,
|
| 100 |
+
"loss": 3.8597,
|
| 101 |
+
"step": 325
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.04427017455097394,
|
| 105 |
+
"grad_norm": 1.2955825328826904,
|
| 106 |
+
"learning_rate": 1.911712623324058e-05,
|
| 107 |
+
"loss": 3.6891,
|
| 108 |
+
"step": 350
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.04743232987604351,
|
| 112 |
+
"grad_norm": 1.9034608602523804,
|
| 113 |
+
"learning_rate": 1.905388312673919e-05,
|
| 114 |
+
"loss": 3.8955,
|
| 115 |
+
"step": 375
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.05059448520111308,
|
| 119 |
+
"grad_norm": 1.649466872215271,
|
| 120 |
+
"learning_rate": 1.8990640020237796e-05,
|
| 121 |
+
"loss": 3.8186,
|
| 122 |
+
"step": 400
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.05375664052618265,
|
| 126 |
+
"grad_norm": 1.8459181785583496,
|
| 127 |
+
"learning_rate": 1.8927396913736406e-05,
|
| 128 |
+
"loss": 3.5247,
|
| 129 |
+
"step": 425
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.056918795851252216,
|
| 133 |
+
"grad_norm": 1.5839076042175293,
|
| 134 |
+
"learning_rate": 1.8864153807235012e-05,
|
| 135 |
+
"loss": 3.756,
|
| 136 |
+
"step": 450
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.060080951176321784,
|
| 140 |
+
"grad_norm": 1.5614947080612183,
|
| 141 |
+
"learning_rate": 1.8800910700733622e-05,
|
| 142 |
+
"loss": 3.6433,
|
| 143 |
+
"step": 475
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.06324310650139135,
|
| 147 |
+
"grad_norm": 1.7831311225891113,
|
| 148 |
+
"learning_rate": 1.873766759423223e-05,
|
| 149 |
+
"loss": 3.9025,
|
| 150 |
+
"step": 500
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.06640526182646092,
|
| 154 |
+
"grad_norm": 1.4928420782089233,
|
| 155 |
+
"learning_rate": 1.867442448773084e-05,
|
| 156 |
+
"loss": 3.4216,
|
| 157 |
+
"step": 525
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.06956741715153049,
|
| 161 |
+
"grad_norm": 1.8199015855789185,
|
| 162 |
+
"learning_rate": 1.861118138122945e-05,
|
| 163 |
+
"loss": 3.7497,
|
| 164 |
+
"step": 550
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.07272957247660006,
|
| 168 |
+
"grad_norm": 1.6205722093582153,
|
| 169 |
+
"learning_rate": 1.854793827472806e-05,
|
| 170 |
+
"loss": 3.4153,
|
| 171 |
+
"step": 575
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.07589172780166961,
|
| 175 |
+
"grad_norm": 1.4476927518844604,
|
| 176 |
+
"learning_rate": 1.8484695168226665e-05,
|
| 177 |
+
"loss": 3.6298,
|
| 178 |
+
"step": 600
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.07905388312673918,
|
| 182 |
+
"grad_norm": 1.5358431339263916,
|
| 183 |
+
"learning_rate": 1.8421452061725275e-05,
|
| 184 |
+
"loss": 3.5282,
|
| 185 |
+
"step": 625
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.08221603845180875,
|
| 189 |
+
"grad_norm": 1.616495966911316,
|
| 190 |
+
"learning_rate": 1.835820895522388e-05,
|
| 191 |
+
"loss": 3.6029,
|
| 192 |
+
"step": 650
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.08537819377687832,
|
| 196 |
+
"grad_norm": 1.4835790395736694,
|
| 197 |
+
"learning_rate": 1.829496584872249e-05,
|
| 198 |
+
"loss": 3.6817,
|
| 199 |
+
"step": 675
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.08854034910194788,
|
| 203 |
+
"grad_norm": 1.8951404094696045,
|
| 204 |
+
"learning_rate": 1.8231722742221098e-05,
|
| 205 |
+
"loss": 3.5222,
|
| 206 |
+
"step": 700
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.09170250442701745,
|
| 210 |
+
"grad_norm": 1.4792296886444092,
|
| 211 |
+
"learning_rate": 1.8168479635719707e-05,
|
| 212 |
+
"loss": 3.552,
|
| 213 |
+
"step": 725
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.09486465975208702,
|
| 217 |
+
"grad_norm": 1.5331674814224243,
|
| 218 |
+
"learning_rate": 1.8105236529218317e-05,
|
| 219 |
+
"loss": 3.6274,
|
| 220 |
+
"step": 750
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.09802681507715659,
|
| 224 |
+
"grad_norm": 1.3521767854690552,
|
| 225 |
+
"learning_rate": 1.8041993422716927e-05,
|
| 226 |
+
"loss": 3.4604,
|
| 227 |
+
"step": 775
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.10118897040222616,
|
| 231 |
+
"grad_norm": 1.9199748039245605,
|
| 232 |
+
"learning_rate": 1.7978750316215534e-05,
|
| 233 |
+
"loss": 3.6434,
|
| 234 |
+
"step": 800
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.10435112572729573,
|
| 238 |
+
"grad_norm": 1.4599398374557495,
|
| 239 |
+
"learning_rate": 1.7915507209714144e-05,
|
| 240 |
+
"loss": 3.4381,
|
| 241 |
+
"step": 825
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.1075132810523653,
|
| 245 |
+
"grad_norm": 1.8845282793045044,
|
| 246 |
+
"learning_rate": 1.785226410321275e-05,
|
| 247 |
+
"loss": 3.4885,
|
| 248 |
+
"step": 850
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.11067543637743486,
|
| 252 |
+
"grad_norm": 1.6054080724716187,
|
| 253 |
+
"learning_rate": 1.778902099671136e-05,
|
| 254 |
+
"loss": 3.4714,
|
| 255 |
+
"step": 875
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.11383759170250443,
|
| 259 |
+
"grad_norm": 1.4243124723434448,
|
| 260 |
+
"learning_rate": 1.7725777890209967e-05,
|
| 261 |
+
"loss": 3.423,
|
| 262 |
+
"step": 900
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.116999747027574,
|
| 266 |
+
"grad_norm": 1.828045129776001,
|
| 267 |
+
"learning_rate": 1.7662534783708576e-05,
|
| 268 |
+
"loss": 3.3451,
|
| 269 |
+
"step": 925
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.12016190235264357,
|
| 273 |
+
"grad_norm": 1.7184784412384033,
|
| 274 |
+
"learning_rate": 1.7599291677207186e-05,
|
| 275 |
+
"loss": 3.4567,
|
| 276 |
+
"step": 950
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.12332405767771312,
|
| 280 |
+
"grad_norm": 2.0031352043151855,
|
| 281 |
+
"learning_rate": 1.7536048570705796e-05,
|
| 282 |
+
"loss": 3.5227,
|
| 283 |
+
"step": 975
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.1264862130027827,
|
| 287 |
+
"grad_norm": 1.8211214542388916,
|
| 288 |
+
"learning_rate": 1.7472805464204403e-05,
|
| 289 |
+
"loss": 3.4222,
|
| 290 |
+
"step": 1000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.12964836832785226,
|
| 294 |
+
"grad_norm": 1.6366448402404785,
|
| 295 |
+
"learning_rate": 1.7409562357703013e-05,
|
| 296 |
+
"loss": 3.3479,
|
| 297 |
+
"step": 1025
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.13281052365292184,
|
| 301 |
+
"grad_norm": 1.4770481586456299,
|
| 302 |
+
"learning_rate": 1.734631925120162e-05,
|
| 303 |
+
"loss": 3.4232,
|
| 304 |
+
"step": 1050
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.1359726789779914,
|
| 308 |
+
"grad_norm": 1.3028419017791748,
|
| 309 |
+
"learning_rate": 1.728307614470023e-05,
|
| 310 |
+
"loss": 3.2511,
|
| 311 |
+
"step": 1075
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.13913483430306098,
|
| 315 |
+
"grad_norm": 1.8425815105438232,
|
| 316 |
+
"learning_rate": 1.721983303819884e-05,
|
| 317 |
+
"loss": 3.3851,
|
| 318 |
+
"step": 1100
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.14229698962813053,
|
| 322 |
+
"grad_norm": 1.5002127885818481,
|
| 323 |
+
"learning_rate": 1.715658993169745e-05,
|
| 324 |
+
"loss": 3.5281,
|
| 325 |
+
"step": 1125
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.14545914495320011,
|
| 329 |
+
"grad_norm": 1.5222417116165161,
|
| 330 |
+
"learning_rate": 1.7093346825196055e-05,
|
| 331 |
+
"loss": 3.3473,
|
| 332 |
+
"step": 1150
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.14862130027826967,
|
| 336 |
+
"grad_norm": 1.628788948059082,
|
| 337 |
+
"learning_rate": 1.7030103718694665e-05,
|
| 338 |
+
"loss": 3.3795,
|
| 339 |
+
"step": 1175
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.15178345560333922,
|
| 343 |
+
"grad_norm": 1.3158140182495117,
|
| 344 |
+
"learning_rate": 1.6966860612193272e-05,
|
| 345 |
+
"loss": 3.5536,
|
| 346 |
+
"step": 1200
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.1549456109284088,
|
| 350 |
+
"grad_norm": 1.8109098672866821,
|
| 351 |
+
"learning_rate": 1.690361750569188e-05,
|
| 352 |
+
"loss": 3.3006,
|
| 353 |
+
"step": 1225
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.15810776625347836,
|
| 357 |
+
"grad_norm": 2.0179309844970703,
|
| 358 |
+
"learning_rate": 1.6840374399190488e-05,
|
| 359 |
+
"loss": 3.4737,
|
| 360 |
+
"step": 1250
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.16126992157854794,
|
| 364 |
+
"grad_norm": 1.8066917657852173,
|
| 365 |
+
"learning_rate": 1.6777131292689098e-05,
|
| 366 |
+
"loss": 3.3135,
|
| 367 |
+
"step": 1275
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.1644320769036175,
|
| 371 |
+
"grad_norm": 1.4508979320526123,
|
| 372 |
+
"learning_rate": 1.6713888186187708e-05,
|
| 373 |
+
"loss": 3.154,
|
| 374 |
+
"step": 1300
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.16759423222868708,
|
| 378 |
+
"grad_norm": 1.324156641960144,
|
| 379 |
+
"learning_rate": 1.6650645079686318e-05,
|
| 380 |
+
"loss": 3.3076,
|
| 381 |
+
"step": 1325
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.17075638755375663,
|
| 385 |
+
"grad_norm": 1.7277374267578125,
|
| 386 |
+
"learning_rate": 1.6587401973184924e-05,
|
| 387 |
+
"loss": 3.3168,
|
| 388 |
+
"step": 1350
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.17391854287882622,
|
| 392 |
+
"grad_norm": 1.581597924232483,
|
| 393 |
+
"learning_rate": 1.6524158866683534e-05,
|
| 394 |
+
"loss": 3.2586,
|
| 395 |
+
"step": 1375
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.17708069820389577,
|
| 399 |
+
"grad_norm": 1.2408771514892578,
|
| 400 |
+
"learning_rate": 1.646091576018214e-05,
|
| 401 |
+
"loss": 3.3174,
|
| 402 |
+
"step": 1400
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.18024285352896535,
|
| 406 |
+
"grad_norm": 1.5069456100463867,
|
| 407 |
+
"learning_rate": 1.639767265368075e-05,
|
| 408 |
+
"loss": 3.2248,
|
| 409 |
+
"step": 1425
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.1834050088540349,
|
| 413 |
+
"grad_norm": 1.474717378616333,
|
| 414 |
+
"learning_rate": 1.6334429547179357e-05,
|
| 415 |
+
"loss": 3.2757,
|
| 416 |
+
"step": 1450
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.1865671641791045,
|
| 420 |
+
"grad_norm": 1.648443341255188,
|
| 421 |
+
"learning_rate": 1.6271186440677967e-05,
|
| 422 |
+
"loss": 3.2754,
|
| 423 |
+
"step": 1475
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.18972931950417404,
|
| 427 |
+
"grad_norm": 1.5270763635635376,
|
| 428 |
+
"learning_rate": 1.6207943334176577e-05,
|
| 429 |
+
"loss": 3.2413,
|
| 430 |
+
"step": 1500
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.19289147482924363,
|
| 434 |
+
"grad_norm": 1.643730640411377,
|
| 435 |
+
"learning_rate": 1.6144700227675187e-05,
|
| 436 |
+
"loss": 3.213,
|
| 437 |
+
"step": 1525
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.19605363015431318,
|
| 441 |
+
"grad_norm": 1.4000329971313477,
|
| 442 |
+
"learning_rate": 1.6081457121173793e-05,
|
| 443 |
+
"loss": 3.236,
|
| 444 |
+
"step": 1550
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.19921578547938273,
|
| 448 |
+
"grad_norm": 1.6957166194915771,
|
| 449 |
+
"learning_rate": 1.6018214014672403e-05,
|
| 450 |
+
"loss": 3.1811,
|
| 451 |
+
"step": 1575
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.20237794080445232,
|
| 455 |
+
"grad_norm": 1.763476848602295,
|
| 456 |
+
"learning_rate": 1.595497090817101e-05,
|
| 457 |
+
"loss": 3.2634,
|
| 458 |
+
"step": 1600
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.20554009612952187,
|
| 462 |
+
"grad_norm": 1.5622406005859375,
|
| 463 |
+
"learning_rate": 1.589172780166962e-05,
|
| 464 |
+
"loss": 3.2498,
|
| 465 |
+
"step": 1625
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.20870225145459145,
|
| 469 |
+
"grad_norm": 2.501917600631714,
|
| 470 |
+
"learning_rate": 1.5828484695168226e-05,
|
| 471 |
+
"loss": 3.3526,
|
| 472 |
+
"step": 1650
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.211864406779661,
|
| 476 |
+
"grad_norm": 1.5445410013198853,
|
| 477 |
+
"learning_rate": 1.5765241588666836e-05,
|
| 478 |
+
"loss": 3.25,
|
| 479 |
+
"step": 1675
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.2150265621047306,
|
| 483 |
+
"grad_norm": 1.420432209968567,
|
| 484 |
+
"learning_rate": 1.5701998482165446e-05,
|
| 485 |
+
"loss": 3.316,
|
| 486 |
+
"step": 1700
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.21818871742980014,
|
| 490 |
+
"grad_norm": 1.7002512216567993,
|
| 491 |
+
"learning_rate": 1.5638755375664056e-05,
|
| 492 |
+
"loss": 3.156,
|
| 493 |
+
"step": 1725
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.22135087275486973,
|
| 497 |
+
"grad_norm": 1.533703088760376,
|
| 498 |
+
"learning_rate": 1.5575512269162662e-05,
|
| 499 |
+
"loss": 3.0938,
|
| 500 |
+
"step": 1750
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.22451302807993928,
|
| 504 |
+
"grad_norm": 1.4791340827941895,
|
| 505 |
+
"learning_rate": 1.5512269162661272e-05,
|
| 506 |
+
"loss": 3.2756,
|
| 507 |
+
"step": 1775
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.22767518340500886,
|
| 511 |
+
"grad_norm": 1.783389925956726,
|
| 512 |
+
"learning_rate": 1.544902605615988e-05,
|
| 513 |
+
"loss": 3.3588,
|
| 514 |
+
"step": 1800
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.23083733873007842,
|
| 518 |
+
"grad_norm": 1.7652767896652222,
|
| 519 |
+
"learning_rate": 1.538578294965849e-05,
|
| 520 |
+
"loss": 3.1746,
|
| 521 |
+
"step": 1825
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.233999494055148,
|
| 525 |
+
"grad_norm": 1.7373157739639282,
|
| 526 |
+
"learning_rate": 1.53225398431571e-05,
|
| 527 |
+
"loss": 3.2757,
|
| 528 |
+
"step": 1850
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.23716164938021755,
|
| 532 |
+
"grad_norm": 1.655953049659729,
|
| 533 |
+
"learning_rate": 1.5259296736655705e-05,
|
| 534 |
+
"loss": 3.2366,
|
| 535 |
+
"step": 1875
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.24032380470528714,
|
| 539 |
+
"grad_norm": 1.4808825254440308,
|
| 540 |
+
"learning_rate": 1.5196053630154315e-05,
|
| 541 |
+
"loss": 3.2056,
|
| 542 |
+
"step": 1900
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.2434859600303567,
|
| 546 |
+
"grad_norm": 1.3337376117706299,
|
| 547 |
+
"learning_rate": 1.5132810523652923e-05,
|
| 548 |
+
"loss": 2.9247,
|
| 549 |
+
"step": 1925
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.24664811535542625,
|
| 553 |
+
"grad_norm": 1.3115172386169434,
|
| 554 |
+
"learning_rate": 1.5069567417151531e-05,
|
| 555 |
+
"loss": 3.3964,
|
| 556 |
+
"step": 1950
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.24981027068049583,
|
| 560 |
+
"grad_norm": 1.3567864894866943,
|
| 561 |
+
"learning_rate": 1.500632431065014e-05,
|
| 562 |
+
"loss": 3.2777,
|
| 563 |
+
"step": 1975
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.2529724260055654,
|
| 567 |
+
"grad_norm": 1.5610828399658203,
|
| 568 |
+
"learning_rate": 1.4943081204148748e-05,
|
| 569 |
+
"loss": 3.0115,
|
| 570 |
+
"step": 2000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.25613458133063494,
|
| 574 |
+
"grad_norm": 1.6952314376831055,
|
| 575 |
+
"learning_rate": 1.4879838097647357e-05,
|
| 576 |
+
"loss": 3.2321,
|
| 577 |
+
"step": 2025
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.2592967366557045,
|
| 581 |
+
"grad_norm": 1.5469592809677124,
|
| 582 |
+
"learning_rate": 1.4816594991145967e-05,
|
| 583 |
+
"loss": 3.1697,
|
| 584 |
+
"step": 2050
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.2624588919807741,
|
| 588 |
+
"grad_norm": 1.7456170320510864,
|
| 589 |
+
"learning_rate": 1.4753351884644576e-05,
|
| 590 |
+
"loss": 3.0305,
|
| 591 |
+
"step": 2075
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.2656210473058437,
|
| 595 |
+
"grad_norm": 1.8613225221633911,
|
| 596 |
+
"learning_rate": 1.4690108778143184e-05,
|
| 597 |
+
"loss": 3.1094,
|
| 598 |
+
"step": 2100
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.2687832026309132,
|
| 602 |
+
"grad_norm": 1.6130377054214478,
|
| 603 |
+
"learning_rate": 1.4626865671641792e-05,
|
| 604 |
+
"loss": 3.1656,
|
| 605 |
+
"step": 2125
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.2719453579559828,
|
| 609 |
+
"grad_norm": 1.4752004146575928,
|
| 610 |
+
"learning_rate": 1.45636225651404e-05,
|
| 611 |
+
"loss": 3.1527,
|
| 612 |
+
"step": 2150
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.2751075132810524,
|
| 616 |
+
"grad_norm": 1.3546524047851562,
|
| 617 |
+
"learning_rate": 1.4500379458639008e-05,
|
| 618 |
+
"loss": 3.1096,
|
| 619 |
+
"step": 2175
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.27826966860612196,
|
| 623 |
+
"grad_norm": 1.5383223295211792,
|
| 624 |
+
"learning_rate": 1.4437136352137617e-05,
|
| 625 |
+
"loss": 3.3534,
|
| 626 |
+
"step": 2200
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.2814318239311915,
|
| 630 |
+
"grad_norm": 1.7905950546264648,
|
| 631 |
+
"learning_rate": 1.4373893245636228e-05,
|
| 632 |
+
"loss": 3.2779,
|
| 633 |
+
"step": 2225
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.28459397925626106,
|
| 637 |
+
"grad_norm": 1.408922553062439,
|
| 638 |
+
"learning_rate": 1.4310650139134836e-05,
|
| 639 |
+
"loss": 3.0517,
|
| 640 |
+
"step": 2250
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.28775613458133065,
|
| 644 |
+
"grad_norm": 1.7881851196289062,
|
| 645 |
+
"learning_rate": 1.4247407032633445e-05,
|
| 646 |
+
"loss": 3.144,
|
| 647 |
+
"step": 2275
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.29091828990640023,
|
| 651 |
+
"grad_norm": 1.5622018575668335,
|
| 652 |
+
"learning_rate": 1.4184163926132053e-05,
|
| 653 |
+
"loss": 3.0161,
|
| 654 |
+
"step": 2300
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.29408044523146976,
|
| 658 |
+
"grad_norm": 1.1805533170700073,
|
| 659 |
+
"learning_rate": 1.4120920819630661e-05,
|
| 660 |
+
"loss": 3.3535,
|
| 661 |
+
"step": 2325
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.29724260055653934,
|
| 665 |
+
"grad_norm": 1.4643152952194214,
|
| 666 |
+
"learning_rate": 1.4057677713129269e-05,
|
| 667 |
+
"loss": 3.0817,
|
| 668 |
+
"step": 2350
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.3004047558816089,
|
| 672 |
+
"grad_norm": 1.640499234199524,
|
| 673 |
+
"learning_rate": 1.3994434606627877e-05,
|
| 674 |
+
"loss": 3.2889,
|
| 675 |
+
"step": 2375
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.30356691120667845,
|
| 679 |
+
"grad_norm": 1.4501370191574097,
|
| 680 |
+
"learning_rate": 1.3931191500126486e-05,
|
| 681 |
+
"loss": 3.0626,
|
| 682 |
+
"step": 2400
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.30672906653174803,
|
| 686 |
+
"grad_norm": 1.540034532546997,
|
| 687 |
+
"learning_rate": 1.3867948393625097e-05,
|
| 688 |
+
"loss": 3.1131,
|
| 689 |
+
"step": 2425
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.3098912218568176,
|
| 693 |
+
"grad_norm": 1.3436205387115479,
|
| 694 |
+
"learning_rate": 1.3804705287123705e-05,
|
| 695 |
+
"loss": 3.3248,
|
| 696 |
+
"step": 2450
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.3130533771818872,
|
| 700 |
+
"grad_norm": 1.8110958337783813,
|
| 701 |
+
"learning_rate": 1.3741462180622314e-05,
|
| 702 |
+
"loss": 3.2063,
|
| 703 |
+
"step": 2475
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.3162155325069567,
|
| 707 |
+
"grad_norm": 1.1593376398086548,
|
| 708 |
+
"learning_rate": 1.3678219074120922e-05,
|
| 709 |
+
"loss": 3.1707,
|
| 710 |
+
"step": 2500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.3193776878320263,
|
| 714 |
+
"grad_norm": 1.2828359603881836,
|
| 715 |
+
"learning_rate": 1.361497596761953e-05,
|
| 716 |
+
"loss": 3.1273,
|
| 717 |
+
"step": 2525
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.3225398431570959,
|
| 721 |
+
"grad_norm": 1.4801135063171387,
|
| 722 |
+
"learning_rate": 1.3551732861118138e-05,
|
| 723 |
+
"loss": 3.1145,
|
| 724 |
+
"step": 2550
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.32570199848216547,
|
| 728 |
+
"grad_norm": 1.6385855674743652,
|
| 729 |
+
"learning_rate": 1.3488489754616746e-05,
|
| 730 |
+
"loss": 3.1274,
|
| 731 |
+
"step": 2575
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.328864153807235,
|
| 735 |
+
"grad_norm": 1.3759686946868896,
|
| 736 |
+
"learning_rate": 1.3425246648115358e-05,
|
| 737 |
+
"loss": 2.9515,
|
| 738 |
+
"step": 2600
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.3320263091323046,
|
| 742 |
+
"grad_norm": 1.3556058406829834,
|
| 743 |
+
"learning_rate": 1.3362003541613966e-05,
|
| 744 |
+
"loss": 3.1251,
|
| 745 |
+
"step": 2625
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.33518846445737416,
|
| 749 |
+
"grad_norm": 1.8028017282485962,
|
| 750 |
+
"learning_rate": 1.3298760435112574e-05,
|
| 751 |
+
"loss": 3.1851,
|
| 752 |
+
"step": 2650
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.33835061978244374,
|
| 756 |
+
"grad_norm": 1.3875707387924194,
|
| 757 |
+
"learning_rate": 1.3235517328611182e-05,
|
| 758 |
+
"loss": 3.1831,
|
| 759 |
+
"step": 2675
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.34151277510751327,
|
| 763 |
+
"grad_norm": 1.3739882707595825,
|
| 764 |
+
"learning_rate": 1.317227422210979e-05,
|
| 765 |
+
"loss": 2.944,
|
| 766 |
+
"step": 2700
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.34467493043258285,
|
| 770 |
+
"grad_norm": 1.7388805150985718,
|
| 771 |
+
"learning_rate": 1.3109031115608399e-05,
|
| 772 |
+
"loss": 2.9621,
|
| 773 |
+
"step": 2725
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.34783708575765243,
|
| 777 |
+
"grad_norm": 1.4218101501464844,
|
| 778 |
+
"learning_rate": 1.3045788009107007e-05,
|
| 779 |
+
"loss": 3.1734,
|
| 780 |
+
"step": 2750
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.35099924108272196,
|
| 784 |
+
"grad_norm": 1.5523903369903564,
|
| 785 |
+
"learning_rate": 1.2982544902605615e-05,
|
| 786 |
+
"loss": 3.1323,
|
| 787 |
+
"step": 2775
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.35416139640779154,
|
| 791 |
+
"grad_norm": 1.2134604454040527,
|
| 792 |
+
"learning_rate": 1.2919301796104227e-05,
|
| 793 |
+
"loss": 2.9738,
|
| 794 |
+
"step": 2800
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.3573235517328611,
|
| 798 |
+
"grad_norm": 1.7392768859863281,
|
| 799 |
+
"learning_rate": 1.2856058689602835e-05,
|
| 800 |
+
"loss": 3.2412,
|
| 801 |
+
"step": 2825
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.3604857070579307,
|
| 805 |
+
"grad_norm": 1.5656300783157349,
|
| 806 |
+
"learning_rate": 1.2792815583101443e-05,
|
| 807 |
+
"loss": 3.1549,
|
| 808 |
+
"step": 2850
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.36364786238300023,
|
| 812 |
+
"grad_norm": 1.592602014541626,
|
| 813 |
+
"learning_rate": 1.2729572476600051e-05,
|
| 814 |
+
"loss": 3.2068,
|
| 815 |
+
"step": 2875
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.3668100177080698,
|
| 819 |
+
"grad_norm": 1.6737302541732788,
|
| 820 |
+
"learning_rate": 1.266632937009866e-05,
|
| 821 |
+
"loss": 3.0918,
|
| 822 |
+
"step": 2900
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.3699721730331394,
|
| 826 |
+
"grad_norm": 1.4579912424087524,
|
| 827 |
+
"learning_rate": 1.2603086263597268e-05,
|
| 828 |
+
"loss": 3.17,
|
| 829 |
+
"step": 2925
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.373134328358209,
|
| 833 |
+
"grad_norm": 1.9524872303009033,
|
| 834 |
+
"learning_rate": 1.2539843157095876e-05,
|
| 835 |
+
"loss": 3.0157,
|
| 836 |
+
"step": 2950
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.3762964836832785,
|
| 840 |
+
"grad_norm": 1.5913100242614746,
|
| 841 |
+
"learning_rate": 1.2476600050594488e-05,
|
| 842 |
+
"loss": 2.9347,
|
| 843 |
+
"step": 2975
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.3794586390083481,
|
| 847 |
+
"grad_norm": 1.5421273708343506,
|
| 848 |
+
"learning_rate": 1.2413356944093096e-05,
|
| 849 |
+
"loss": 3.0643,
|
| 850 |
+
"step": 3000
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.38262079433341767,
|
| 854 |
+
"grad_norm": 1.4578076601028442,
|
| 855 |
+
"learning_rate": 1.2350113837591704e-05,
|
| 856 |
+
"loss": 3.1577,
|
| 857 |
+
"step": 3025
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.38578294965848725,
|
| 861 |
+
"grad_norm": 1.4034929275512695,
|
| 862 |
+
"learning_rate": 1.2286870731090312e-05,
|
| 863 |
+
"loss": 3.0522,
|
| 864 |
+
"step": 3050
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.3889451049835568,
|
| 868 |
+
"grad_norm": 1.4334189891815186,
|
| 869 |
+
"learning_rate": 1.222362762458892e-05,
|
| 870 |
+
"loss": 3.0195,
|
| 871 |
+
"step": 3075
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.39210726030862636,
|
| 875 |
+
"grad_norm": 1.5584102869033813,
|
| 876 |
+
"learning_rate": 1.2160384518087529e-05,
|
| 877 |
+
"loss": 2.9582,
|
| 878 |
+
"step": 3100
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.39526941563369594,
|
| 882 |
+
"grad_norm": 1.4117885828018188,
|
| 883 |
+
"learning_rate": 1.2097141411586137e-05,
|
| 884 |
+
"loss": 3.1979,
|
| 885 |
+
"step": 3125
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.39843157095876547,
|
| 889 |
+
"grad_norm": 1.6135624647140503,
|
| 890 |
+
"learning_rate": 1.2033898305084745e-05,
|
| 891 |
+
"loss": 3.0185,
|
| 892 |
+
"step": 3150
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.40159372628383505,
|
| 896 |
+
"grad_norm": 1.9495983123779297,
|
| 897 |
+
"learning_rate": 1.1970655198583357e-05,
|
| 898 |
+
"loss": 3.1753,
|
| 899 |
+
"step": 3175
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.40475588160890463,
|
| 903 |
+
"grad_norm": 1.3732932806015015,
|
| 904 |
+
"learning_rate": 1.1907412092081965e-05,
|
| 905 |
+
"loss": 2.9982,
|
| 906 |
+
"step": 3200
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.4079180369339742,
|
| 910 |
+
"grad_norm": 1.6417973041534424,
|
| 911 |
+
"learning_rate": 1.1844168985580573e-05,
|
| 912 |
+
"loss": 2.9491,
|
| 913 |
+
"step": 3225
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.41108019225904374,
|
| 917 |
+
"grad_norm": 1.599778413772583,
|
| 918 |
+
"learning_rate": 1.1780925879079181e-05,
|
| 919 |
+
"loss": 2.9619,
|
| 920 |
+
"step": 3250
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.4142423475841133,
|
| 924 |
+
"grad_norm": 1.5678138732910156,
|
| 925 |
+
"learning_rate": 1.171768277257779e-05,
|
| 926 |
+
"loss": 3.1179,
|
| 927 |
+
"step": 3275
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.4174045029091829,
|
| 931 |
+
"grad_norm": 1.4923375844955444,
|
| 932 |
+
"learning_rate": 1.1654439666076398e-05,
|
| 933 |
+
"loss": 3.193,
|
| 934 |
+
"step": 3300
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.4205666582342525,
|
| 938 |
+
"grad_norm": 1.5655521154403687,
|
| 939 |
+
"learning_rate": 1.1591196559575006e-05,
|
| 940 |
+
"loss": 3.0813,
|
| 941 |
+
"step": 3325
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.423728813559322,
|
| 945 |
+
"grad_norm": 1.7306265830993652,
|
| 946 |
+
"learning_rate": 1.1527953453073617e-05,
|
| 947 |
+
"loss": 3.1584,
|
| 948 |
+
"step": 3350
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.4268909688843916,
|
| 952 |
+
"grad_norm": 1.4964895248413086,
|
| 953 |
+
"learning_rate": 1.1464710346572226e-05,
|
| 954 |
+
"loss": 2.9885,
|
| 955 |
+
"step": 3375
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.4300531242094612,
|
| 959 |
+
"grad_norm": 1.386781930923462,
|
| 960 |
+
"learning_rate": 1.1401467240070834e-05,
|
| 961 |
+
"loss": 2.7611,
|
| 962 |
+
"step": 3400
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.43321527953453076,
|
| 966 |
+
"grad_norm": 1.223809838294983,
|
| 967 |
+
"learning_rate": 1.1338224133569442e-05,
|
| 968 |
+
"loss": 2.9793,
|
| 969 |
+
"step": 3425
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.4363774348596003,
|
| 973 |
+
"grad_norm": 1.7949477434158325,
|
| 974 |
+
"learning_rate": 1.127498102706805e-05,
|
| 975 |
+
"loss": 3.1053,
|
| 976 |
+
"step": 3450
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.43953959018466987,
|
| 980 |
+
"grad_norm": 1.2614104747772217,
|
| 981 |
+
"learning_rate": 1.1211737920566658e-05,
|
| 982 |
+
"loss": 3.1497,
|
| 983 |
+
"step": 3475
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.44270174550973945,
|
| 987 |
+
"grad_norm": 1.4934375286102295,
|
| 988 |
+
"learning_rate": 1.1148494814065267e-05,
|
| 989 |
+
"loss": 3.2349,
|
| 990 |
+
"step": 3500
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.445863900834809,
|
| 994 |
+
"grad_norm": 1.692384958267212,
|
| 995 |
+
"learning_rate": 1.1085251707563875e-05,
|
| 996 |
+
"loss": 3.2143,
|
| 997 |
+
"step": 3525
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.44902605615987856,
|
| 1001 |
+
"grad_norm": 1.430724024772644,
|
| 1002 |
+
"learning_rate": 1.1022008601062486e-05,
|
| 1003 |
+
"loss": 2.9926,
|
| 1004 |
+
"step": 3550
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.45218821148494814,
|
| 1008 |
+
"grad_norm": 1.6542253494262695,
|
| 1009 |
+
"learning_rate": 1.0958765494561095e-05,
|
| 1010 |
+
"loss": 3.0877,
|
| 1011 |
+
"step": 3575
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.4553503668100177,
|
| 1015 |
+
"grad_norm": 1.761777400970459,
|
| 1016 |
+
"learning_rate": 1.0895522388059703e-05,
|
| 1017 |
+
"loss": 3.1343,
|
| 1018 |
+
"step": 3600
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.45851252213508725,
|
| 1022 |
+
"grad_norm": 1.3432550430297852,
|
| 1023 |
+
"learning_rate": 1.0832279281558311e-05,
|
| 1024 |
+
"loss": 2.9125,
|
| 1025 |
+
"step": 3625
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.46167467746015683,
|
| 1029 |
+
"grad_norm": 1.6349095106124878,
|
| 1030 |
+
"learning_rate": 1.0769036175056919e-05,
|
| 1031 |
+
"loss": 3.0102,
|
| 1032 |
+
"step": 3650
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.4648368327852264,
|
| 1036 |
+
"grad_norm": 1.5871011018753052,
|
| 1037 |
+
"learning_rate": 1.0705793068555527e-05,
|
| 1038 |
+
"loss": 2.8361,
|
| 1039 |
+
"step": 3675
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.467998988110296,
|
| 1043 |
+
"grad_norm": 1.5581876039505005,
|
| 1044 |
+
"learning_rate": 1.0642549962054136e-05,
|
| 1045 |
+
"loss": 2.9176,
|
| 1046 |
+
"step": 3700
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.4711611434353655,
|
| 1050 |
+
"grad_norm": 1.9018642902374268,
|
| 1051 |
+
"learning_rate": 1.0579306855552747e-05,
|
| 1052 |
+
"loss": 3.1236,
|
| 1053 |
+
"step": 3725
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.4743232987604351,
|
| 1057 |
+
"grad_norm": 1.9724278450012207,
|
| 1058 |
+
"learning_rate": 1.0516063749051355e-05,
|
| 1059 |
+
"loss": 3.2517,
|
| 1060 |
+
"step": 3750
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.4774854540855047,
|
| 1064 |
+
"grad_norm": 1.4184868335723877,
|
| 1065 |
+
"learning_rate": 1.0452820642549963e-05,
|
| 1066 |
+
"loss": 2.7384,
|
| 1067 |
+
"step": 3775
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.4806476094105743,
|
| 1071 |
+
"grad_norm": 1.6251132488250732,
|
| 1072 |
+
"learning_rate": 1.0389577536048572e-05,
|
| 1073 |
+
"loss": 2.9232,
|
| 1074 |
+
"step": 3800
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.4838097647356438,
|
| 1078 |
+
"grad_norm": 1.526455283164978,
|
| 1079 |
+
"learning_rate": 1.032633442954718e-05,
|
| 1080 |
+
"loss": 3.0109,
|
| 1081 |
+
"step": 3825
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.4869719200607134,
|
| 1085 |
+
"grad_norm": 1.5482044219970703,
|
| 1086 |
+
"learning_rate": 1.0263091323045788e-05,
|
| 1087 |
+
"loss": 2.9056,
|
| 1088 |
+
"step": 3850
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.49013407538578296,
|
| 1092 |
+
"grad_norm": 1.5261505842208862,
|
| 1093 |
+
"learning_rate": 1.0199848216544396e-05,
|
| 1094 |
+
"loss": 2.8749,
|
| 1095 |
+
"step": 3875
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.4932962307108525,
|
| 1099 |
+
"grad_norm": 1.8025535345077515,
|
| 1100 |
+
"learning_rate": 1.0136605110043004e-05,
|
| 1101 |
+
"loss": 3.1128,
|
| 1102 |
+
"step": 3900
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.4964583860359221,
|
| 1106 |
+
"grad_norm": 1.428043246269226,
|
| 1107 |
+
"learning_rate": 1.0073362003541616e-05,
|
| 1108 |
+
"loss": 3.0411,
|
| 1109 |
+
"step": 3925
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.49962054136099165,
|
| 1113 |
+
"grad_norm": 1.3403562307357788,
|
| 1114 |
+
"learning_rate": 1.0010118897040224e-05,
|
| 1115 |
+
"loss": 2.9253,
|
| 1116 |
+
"step": 3950
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.5027826966860612,
|
| 1120 |
+
"grad_norm": 1.6576727628707886,
|
| 1121 |
+
"learning_rate": 9.946875790538832e-06,
|
| 1122 |
+
"loss": 2.9877,
|
| 1123 |
+
"step": 3975
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.5059448520111308,
|
| 1127 |
+
"grad_norm": 1.8511061668395996,
|
| 1128 |
+
"learning_rate": 9.88363268403744e-06,
|
| 1129 |
+
"loss": 2.919,
|
| 1130 |
+
"step": 4000
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.5091070073362004,
|
| 1134 |
+
"grad_norm": 1.690691590309143,
|
| 1135 |
+
"learning_rate": 9.820389577536049e-06,
|
| 1136 |
+
"loss": 2.7579,
|
| 1137 |
+
"step": 4025
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.5122691626612699,
|
| 1141 |
+
"grad_norm": 1.4383487701416016,
|
| 1142 |
+
"learning_rate": 9.757146471034657e-06,
|
| 1143 |
+
"loss": 2.8421,
|
| 1144 |
+
"step": 4050
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.5154313179863395,
|
| 1148 |
+
"grad_norm": 1.5412685871124268,
|
| 1149 |
+
"learning_rate": 9.693903364533267e-06,
|
| 1150 |
+
"loss": 2.9609,
|
| 1151 |
+
"step": 4075
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.518593473311409,
|
| 1155 |
+
"grad_norm": 1.7865098714828491,
|
| 1156 |
+
"learning_rate": 9.630660258031875e-06,
|
| 1157 |
+
"loss": 2.887,
|
| 1158 |
+
"step": 4100
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.5217556286364786,
|
| 1162 |
+
"grad_norm": 1.4453928470611572,
|
| 1163 |
+
"learning_rate": 9.567417151530483e-06,
|
| 1164 |
+
"loss": 2.9248,
|
| 1165 |
+
"step": 4125
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.5249177839615482,
|
| 1169 |
+
"grad_norm": 1.6136345863342285,
|
| 1170 |
+
"learning_rate": 9.504174045029093e-06,
|
| 1171 |
+
"loss": 3.1814,
|
| 1172 |
+
"step": 4150
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.5280799392866178,
|
| 1176 |
+
"grad_norm": 1.3460476398468018,
|
| 1177 |
+
"learning_rate": 9.440930938527701e-06,
|
| 1178 |
+
"loss": 2.7766,
|
| 1179 |
+
"step": 4175
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.5312420946116874,
|
| 1183 |
+
"grad_norm": 1.6284033060073853,
|
| 1184 |
+
"learning_rate": 9.37768783202631e-06,
|
| 1185 |
+
"loss": 2.9087,
|
| 1186 |
+
"step": 4200
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.534404249936757,
|
| 1190 |
+
"grad_norm": 1.3353221416473389,
|
| 1191 |
+
"learning_rate": 9.314444725524918e-06,
|
| 1192 |
+
"loss": 2.9512,
|
| 1193 |
+
"step": 4225
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.5375664052618264,
|
| 1197 |
+
"grad_norm": 1.9415779113769531,
|
| 1198 |
+
"learning_rate": 9.251201619023528e-06,
|
| 1199 |
+
"loss": 2.7603,
|
| 1200 |
+
"step": 4250
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.540728560586896,
|
| 1204 |
+
"grad_norm": 1.6942236423492432,
|
| 1205 |
+
"learning_rate": 9.187958512522136e-06,
|
| 1206 |
+
"loss": 3.1245,
|
| 1207 |
+
"step": 4275
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.5438907159119656,
|
| 1211 |
+
"grad_norm": 2.2894318103790283,
|
| 1212 |
+
"learning_rate": 9.124715406020744e-06,
|
| 1213 |
+
"loss": 2.9695,
|
| 1214 |
+
"step": 4300
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.5470528712370352,
|
| 1218 |
+
"grad_norm": 1.1514796018600464,
|
| 1219 |
+
"learning_rate": 9.061472299519352e-06,
|
| 1220 |
+
"loss": 3.0047,
|
| 1221 |
+
"step": 4325
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.5502150265621047,
|
| 1225 |
+
"grad_norm": 1.5051628351211548,
|
| 1226 |
+
"learning_rate": 8.998229193017962e-06,
|
| 1227 |
+
"loss": 3.1653,
|
| 1228 |
+
"step": 4350
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.5533771818871743,
|
| 1232 |
+
"grad_norm": 1.4090489149093628,
|
| 1233 |
+
"learning_rate": 8.93498608651657e-06,
|
| 1234 |
+
"loss": 2.9528,
|
| 1235 |
+
"step": 4375
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.5565393372122439,
|
| 1239 |
+
"grad_norm": 1.6863518953323364,
|
| 1240 |
+
"learning_rate": 8.871742980015179e-06,
|
| 1241 |
+
"loss": 2.8309,
|
| 1242 |
+
"step": 4400
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.5597014925373134,
|
| 1246 |
+
"grad_norm": 1.453518271446228,
|
| 1247 |
+
"learning_rate": 8.808499873513787e-06,
|
| 1248 |
+
"loss": 2.8803,
|
| 1249 |
+
"step": 4425
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.562863647862383,
|
| 1253 |
+
"grad_norm": 1.987695336341858,
|
| 1254 |
+
"learning_rate": 8.745256767012397e-06,
|
| 1255 |
+
"loss": 3.001,
|
| 1256 |
+
"step": 4450
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.5660258031874525,
|
| 1260 |
+
"grad_norm": 1.3925899267196655,
|
| 1261 |
+
"learning_rate": 8.682013660511005e-06,
|
| 1262 |
+
"loss": 2.9407,
|
| 1263 |
+
"step": 4475
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.5691879585125221,
|
| 1267 |
+
"grad_norm": 1.802440881729126,
|
| 1268 |
+
"learning_rate": 8.618770554009613e-06,
|
| 1269 |
+
"loss": 3.0096,
|
| 1270 |
+
"step": 4500
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.5723501138375917,
|
| 1274 |
+
"grad_norm": 1.4417994022369385,
|
| 1275 |
+
"learning_rate": 8.555527447508223e-06,
|
| 1276 |
+
"loss": 2.976,
|
| 1277 |
+
"step": 4525
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.5755122691626613,
|
| 1281 |
+
"grad_norm": 1.7502068281173706,
|
| 1282 |
+
"learning_rate": 8.492284341006831e-06,
|
| 1283 |
+
"loss": 2.8715,
|
| 1284 |
+
"step": 4550
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.5786744244877309,
|
| 1288 |
+
"grad_norm": 1.7311025857925415,
|
| 1289 |
+
"learning_rate": 8.42904123450544e-06,
|
| 1290 |
+
"loss": 2.9176,
|
| 1291 |
+
"step": 4575
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.5818365798128005,
|
| 1295 |
+
"grad_norm": 1.6848989725112915,
|
| 1296 |
+
"learning_rate": 8.365798128004048e-06,
|
| 1297 |
+
"loss": 2.9213,
|
| 1298 |
+
"step": 4600
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.5849987351378699,
|
| 1302 |
+
"grad_norm": 1.4339770078659058,
|
| 1303 |
+
"learning_rate": 8.302555021502657e-06,
|
| 1304 |
+
"loss": 2.944,
|
| 1305 |
+
"step": 4625
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.5881608904629395,
|
| 1309 |
+
"grad_norm": 1.4645488262176514,
|
| 1310 |
+
"learning_rate": 8.239311915001266e-06,
|
| 1311 |
+
"loss": 2.8865,
|
| 1312 |
+
"step": 4650
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.5913230457880091,
|
| 1316 |
+
"grad_norm": 1.793530821800232,
|
| 1317 |
+
"learning_rate": 8.176068808499874e-06,
|
| 1318 |
+
"loss": 3.0749,
|
| 1319 |
+
"step": 4675
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.5944852011130787,
|
| 1323 |
+
"grad_norm": 1.753974437713623,
|
| 1324 |
+
"learning_rate": 8.112825701998482e-06,
|
| 1325 |
+
"loss": 2.9768,
|
| 1326 |
+
"step": 4700
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.5976473564381483,
|
| 1330 |
+
"grad_norm": 1.4546597003936768,
|
| 1331 |
+
"learning_rate": 8.049582595497092e-06,
|
| 1332 |
+
"loss": 2.8419,
|
| 1333 |
+
"step": 4725
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.6008095117632178,
|
| 1337 |
+
"grad_norm": 2.1190099716186523,
|
| 1338 |
+
"learning_rate": 7.9863394889957e-06,
|
| 1339 |
+
"loss": 2.8468,
|
| 1340 |
+
"step": 4750
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.6039716670882874,
|
| 1344 |
+
"grad_norm": 2.026573657989502,
|
| 1345 |
+
"learning_rate": 7.923096382494308e-06,
|
| 1346 |
+
"loss": 3.0328,
|
| 1347 |
+
"step": 4775
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.6071338224133569,
|
| 1351 |
+
"grad_norm": 1.5188226699829102,
|
| 1352 |
+
"learning_rate": 7.859853275992917e-06,
|
| 1353 |
+
"loss": 2.7536,
|
| 1354 |
+
"step": 4800
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.6102959777384265,
|
| 1358 |
+
"grad_norm": 1.9499627351760864,
|
| 1359 |
+
"learning_rate": 7.796610169491526e-06,
|
| 1360 |
+
"loss": 3.0033,
|
| 1361 |
+
"step": 4825
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.6134581330634961,
|
| 1365 |
+
"grad_norm": 1.4692327976226807,
|
| 1366 |
+
"learning_rate": 7.733367062990135e-06,
|
| 1367 |
+
"loss": 2.9335,
|
| 1368 |
+
"step": 4850
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.6166202883885656,
|
| 1372 |
+
"grad_norm": 1.4884922504425049,
|
| 1373 |
+
"learning_rate": 7.670123956488743e-06,
|
| 1374 |
+
"loss": 2.9247,
|
| 1375 |
+
"step": 4875
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.6197824437136352,
|
| 1379 |
+
"grad_norm": 1.5420995950698853,
|
| 1380 |
+
"learning_rate": 7.606880849987353e-06,
|
| 1381 |
+
"loss": 2.9574,
|
| 1382 |
+
"step": 4900
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.6229445990387048,
|
| 1386 |
+
"grad_norm": 1.8658238649368286,
|
| 1387 |
+
"learning_rate": 7.543637743485961e-06,
|
| 1388 |
+
"loss": 2.9001,
|
| 1389 |
+
"step": 4925
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.6261067543637744,
|
| 1393 |
+
"grad_norm": 1.5181258916854858,
|
| 1394 |
+
"learning_rate": 7.480394636984569e-06,
|
| 1395 |
+
"loss": 2.7142,
|
| 1396 |
+
"step": 4950
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.629268909688844,
|
| 1400 |
+
"grad_norm": 1.4289120435714722,
|
| 1401 |
+
"learning_rate": 7.417151530483177e-06,
|
| 1402 |
+
"loss": 3.0176,
|
| 1403 |
+
"step": 4975
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.6324310650139134,
|
| 1407 |
+
"grad_norm": 1.7044678926467896,
|
| 1408 |
+
"learning_rate": 7.353908423981787e-06,
|
| 1409 |
+
"loss": 2.9688,
|
| 1410 |
+
"step": 5000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.635593220338983,
|
| 1414 |
+
"grad_norm": 1.2686058282852173,
|
| 1415 |
+
"learning_rate": 7.290665317480395e-06,
|
| 1416 |
+
"loss": 3.0132,
|
| 1417 |
+
"step": 5025
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.6387553756640526,
|
| 1421 |
+
"grad_norm": 1.4413508176803589,
|
| 1422 |
+
"learning_rate": 7.2274222109790036e-06,
|
| 1423 |
+
"loss": 2.948,
|
| 1424 |
+
"step": 5050
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.6419175309891222,
|
| 1428 |
+
"grad_norm": 1.5472356081008911,
|
| 1429 |
+
"learning_rate": 7.164179104477612e-06,
|
| 1430 |
+
"loss": 2.8153,
|
| 1431 |
+
"step": 5075
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.6450796863141918,
|
| 1435 |
+
"grad_norm": 1.3920260667800903,
|
| 1436 |
+
"learning_rate": 7.100935997976222e-06,
|
| 1437 |
+
"loss": 2.9269,
|
| 1438 |
+
"step": 5100
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.6482418416392614,
|
| 1442 |
+
"grad_norm": 1.8862273693084717,
|
| 1443 |
+
"learning_rate": 7.03769289147483e-06,
|
| 1444 |
+
"loss": 2.9641,
|
| 1445 |
+
"step": 5125
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.6514039969643309,
|
| 1449 |
+
"grad_norm": 1.2860732078552246,
|
| 1450 |
+
"learning_rate": 6.974449784973438e-06,
|
| 1451 |
+
"loss": 2.9628,
|
| 1452 |
+
"step": 5150
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.6545661522894004,
|
| 1456 |
+
"grad_norm": 1.6975411176681519,
|
| 1457 |
+
"learning_rate": 6.911206678472046e-06,
|
| 1458 |
+
"loss": 2.8478,
|
| 1459 |
+
"step": 5175
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.65772830761447,
|
| 1463 |
+
"grad_norm": 1.577736258506775,
|
| 1464 |
+
"learning_rate": 6.847963571970656e-06,
|
| 1465 |
+
"loss": 2.8673,
|
| 1466 |
+
"step": 5200
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.6608904629395396,
|
| 1470 |
+
"grad_norm": 1.5640876293182373,
|
| 1471 |
+
"learning_rate": 6.784720465469264e-06,
|
| 1472 |
+
"loss": 2.7907,
|
| 1473 |
+
"step": 5225
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.6640526182646092,
|
| 1477 |
+
"grad_norm": 1.5373408794403076,
|
| 1478 |
+
"learning_rate": 6.7214773589678725e-06,
|
| 1479 |
+
"loss": 2.9308,
|
| 1480 |
+
"step": 5250
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.6672147735896787,
|
| 1484 |
+
"grad_norm": 1.499895453453064,
|
| 1485 |
+
"learning_rate": 6.6582342524664824e-06,
|
| 1486 |
+
"loss": 2.8954,
|
| 1487 |
+
"step": 5275
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.6703769289147483,
|
| 1491 |
+
"grad_norm": 1.7884222269058228,
|
| 1492 |
+
"learning_rate": 6.594991145965091e-06,
|
| 1493 |
+
"loss": 3.0374,
|
| 1494 |
+
"step": 5300
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.6735390842398179,
|
| 1498 |
+
"grad_norm": 1.4689120054244995,
|
| 1499 |
+
"learning_rate": 6.531748039463699e-06,
|
| 1500 |
+
"loss": 2.7937,
|
| 1501 |
+
"step": 5325
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.6767012395648875,
|
| 1505 |
+
"grad_norm": 1.2251571416854858,
|
| 1506 |
+
"learning_rate": 6.468504932962307e-06,
|
| 1507 |
+
"loss": 2.6993,
|
| 1508 |
+
"step": 5350
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.679863394889957,
|
| 1512 |
+
"grad_norm": 1.6900901794433594,
|
| 1513 |
+
"learning_rate": 6.405261826460917e-06,
|
| 1514 |
+
"loss": 2.9683,
|
| 1515 |
+
"step": 5375
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.6830255502150265,
|
| 1519 |
+
"grad_norm": 1.3138439655303955,
|
| 1520 |
+
"learning_rate": 6.342018719959525e-06,
|
| 1521 |
+
"loss": 2.9179,
|
| 1522 |
+
"step": 5400
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.6861877055400961,
|
| 1526 |
+
"grad_norm": 1.3106989860534668,
|
| 1527 |
+
"learning_rate": 6.278775613458133e-06,
|
| 1528 |
+
"loss": 2.8901,
|
| 1529 |
+
"step": 5425
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.6893498608651657,
|
| 1533 |
+
"grad_norm": 1.7244914770126343,
|
| 1534 |
+
"learning_rate": 6.2155325069567415e-06,
|
| 1535 |
+
"loss": 2.8481,
|
| 1536 |
+
"step": 5450
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.6925120161902353,
|
| 1540 |
+
"grad_norm": 1.1945232152938843,
|
| 1541 |
+
"learning_rate": 6.152289400455351e-06,
|
| 1542 |
+
"loss": 2.8619,
|
| 1543 |
+
"step": 5475
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.6956741715153049,
|
| 1547 |
+
"grad_norm": 1.3923369646072388,
|
| 1548 |
+
"learning_rate": 6.08904629395396e-06,
|
| 1549 |
+
"loss": 2.9955,
|
| 1550 |
+
"step": 5500
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.6988363268403744,
|
| 1554 |
+
"grad_norm": 1.7948927879333496,
|
| 1555 |
+
"learning_rate": 6.025803187452568e-06,
|
| 1556 |
+
"loss": 3.0742,
|
| 1557 |
+
"step": 5525
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.7019984821654439,
|
| 1561 |
+
"grad_norm": 1.4123344421386719,
|
| 1562 |
+
"learning_rate": 5.962560080951176e-06,
|
| 1563 |
+
"loss": 2.9499,
|
| 1564 |
+
"step": 5550
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.7051606374905135,
|
| 1568 |
+
"grad_norm": 1.5082272291183472,
|
| 1569 |
+
"learning_rate": 5.899316974449786e-06,
|
| 1570 |
+
"loss": 3.0861,
|
| 1571 |
+
"step": 5575
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.7083227928155831,
|
| 1575 |
+
"grad_norm": 1.6818978786468506,
|
| 1576 |
+
"learning_rate": 5.836073867948394e-06,
|
| 1577 |
+
"loss": 2.7944,
|
| 1578 |
+
"step": 5600
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.7114849481406527,
|
| 1582 |
+
"grad_norm": 2.072303295135498,
|
| 1583 |
+
"learning_rate": 5.772830761447002e-06,
|
| 1584 |
+
"loss": 3.021,
|
| 1585 |
+
"step": 5625
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.7146471034657222,
|
| 1589 |
+
"grad_norm": 1.2416434288024902,
|
| 1590 |
+
"learning_rate": 5.7095876549456105e-06,
|
| 1591 |
+
"loss": 2.7849,
|
| 1592 |
+
"step": 5650
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.7178092587907918,
|
| 1596 |
+
"grad_norm": 1.6787675619125366,
|
| 1597 |
+
"learning_rate": 5.64634454844422e-06,
|
| 1598 |
+
"loss": 2.9102,
|
| 1599 |
+
"step": 5675
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.7209714141158614,
|
| 1603 |
+
"grad_norm": 1.542455792427063,
|
| 1604 |
+
"learning_rate": 5.5831014419428286e-06,
|
| 1605 |
+
"loss": 2.8268,
|
| 1606 |
+
"step": 5700
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.724133569440931,
|
| 1610 |
+
"grad_norm": 2.0236566066741943,
|
| 1611 |
+
"learning_rate": 5.519858335441437e-06,
|
| 1612 |
+
"loss": 2.8222,
|
| 1613 |
+
"step": 5725
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.7272957247660005,
|
| 1617 |
+
"grad_norm": 1.5775691270828247,
|
| 1618 |
+
"learning_rate": 5.456615228940047e-06,
|
| 1619 |
+
"loss": 2.9778,
|
| 1620 |
+
"step": 5750
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.73045788009107,
|
| 1624 |
+
"grad_norm": 1.792724847793579,
|
| 1625 |
+
"learning_rate": 5.393372122438655e-06,
|
| 1626 |
+
"loss": 2.9323,
|
| 1627 |
+
"step": 5775
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.7336200354161396,
|
| 1631 |
+
"grad_norm": 1.5170587301254272,
|
| 1632 |
+
"learning_rate": 5.330129015937263e-06,
|
| 1633 |
+
"loss": 2.8671,
|
| 1634 |
+
"step": 5800
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.7367821907412092,
|
| 1638 |
+
"grad_norm": 1.418395757675171,
|
| 1639 |
+
"learning_rate": 5.266885909435871e-06,
|
| 1640 |
+
"loss": 2.8153,
|
| 1641 |
+
"step": 5825
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.7399443460662788,
|
| 1645 |
+
"grad_norm": 1.4458298683166504,
|
| 1646 |
+
"learning_rate": 5.203642802934481e-06,
|
| 1647 |
+
"loss": 2.9178,
|
| 1648 |
+
"step": 5850
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.7431065013913484,
|
| 1652 |
+
"grad_norm": 1.6139628887176514,
|
| 1653 |
+
"learning_rate": 5.140399696433089e-06,
|
| 1654 |
+
"loss": 2.8857,
|
| 1655 |
+
"step": 5875
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.746268656716418,
|
| 1659 |
+
"grad_norm": 1.1631035804748535,
|
| 1660 |
+
"learning_rate": 5.0771565899316975e-06,
|
| 1661 |
+
"loss": 2.9213,
|
| 1662 |
+
"step": 5900
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.7494308120414874,
|
| 1666 |
+
"grad_norm": 1.5703893899917603,
|
| 1667 |
+
"learning_rate": 5.013913483430306e-06,
|
| 1668 |
+
"loss": 2.891,
|
| 1669 |
+
"step": 5925
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.752592967366557,
|
| 1673 |
+
"grad_norm": 1.4471451044082642,
|
| 1674 |
+
"learning_rate": 4.950670376928915e-06,
|
| 1675 |
+
"loss": 2.8261,
|
| 1676 |
+
"step": 5950
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.7557551226916266,
|
| 1680 |
+
"grad_norm": 1.6305294036865234,
|
| 1681 |
+
"learning_rate": 4.887427270427524e-06,
|
| 1682 |
+
"loss": 2.8918,
|
| 1683 |
+
"step": 5975
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.7589172780166962,
|
| 1687 |
+
"grad_norm": 1.3186579942703247,
|
| 1688 |
+
"learning_rate": 4.824184163926132e-06,
|
| 1689 |
+
"loss": 2.9478,
|
| 1690 |
+
"step": 6000
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.7620794333417658,
|
| 1694 |
+
"grad_norm": 1.4258198738098145,
|
| 1695 |
+
"learning_rate": 4.760941057424741e-06,
|
| 1696 |
+
"loss": 2.9313,
|
| 1697 |
+
"step": 6025
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.7652415886668353,
|
| 1701 |
+
"grad_norm": 1.5036190748214722,
|
| 1702 |
+
"learning_rate": 4.69769795092335e-06,
|
| 1703 |
+
"loss": 2.9562,
|
| 1704 |
+
"step": 6050
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.7684037439919049,
|
| 1708 |
+
"grad_norm": 1.7764675617218018,
|
| 1709 |
+
"learning_rate": 4.634454844421958e-06,
|
| 1710 |
+
"loss": 2.9396,
|
| 1711 |
+
"step": 6075
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.7715658993169745,
|
| 1715 |
+
"grad_norm": 1.624964714050293,
|
| 1716 |
+
"learning_rate": 4.571211737920567e-06,
|
| 1717 |
+
"loss": 2.8789,
|
| 1718 |
+
"step": 6100
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.774728054642044,
|
| 1722 |
+
"grad_norm": 1.1885554790496826,
|
| 1723 |
+
"learning_rate": 4.5079686314191755e-06,
|
| 1724 |
+
"loss": 2.9756,
|
| 1725 |
+
"step": 6125
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.7778902099671136,
|
| 1729 |
+
"grad_norm": 1.752863883972168,
|
| 1730 |
+
"learning_rate": 4.444725524917785e-06,
|
| 1731 |
+
"loss": 2.7279,
|
| 1732 |
+
"step": 6150
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.7810523652921831,
|
| 1736 |
+
"grad_norm": 1.4653725624084473,
|
| 1737 |
+
"learning_rate": 4.381482418416393e-06,
|
| 1738 |
+
"loss": 3.0123,
|
| 1739 |
+
"step": 6175
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.7842145206172527,
|
| 1743 |
+
"grad_norm": 1.5357298851013184,
|
| 1744 |
+
"learning_rate": 4.318239311915002e-06,
|
| 1745 |
+
"loss": 2.7978,
|
| 1746 |
+
"step": 6200
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.7873766759423223,
|
| 1750 |
+
"grad_norm": 1.4677993059158325,
|
| 1751 |
+
"learning_rate": 4.25499620541361e-06,
|
| 1752 |
+
"loss": 2.8016,
|
| 1753 |
+
"step": 6225
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.7905388312673919,
|
| 1757 |
+
"grad_norm": 1.3810391426086426,
|
| 1758 |
+
"learning_rate": 4.191753098912219e-06,
|
| 1759 |
+
"loss": 2.7919,
|
| 1760 |
+
"step": 6250
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.7937009865924615,
|
| 1764 |
+
"grad_norm": 1.827034831047058,
|
| 1765 |
+
"learning_rate": 4.128509992410827e-06,
|
| 1766 |
+
"loss": 2.9552,
|
| 1767 |
+
"step": 6275
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.7968631419175309,
|
| 1771 |
+
"grad_norm": 1.7732640504837036,
|
| 1772 |
+
"learning_rate": 4.065266885909436e-06,
|
| 1773 |
+
"loss": 3.03,
|
| 1774 |
+
"step": 6300
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.8000252972426005,
|
| 1778 |
+
"grad_norm": 2.2995574474334717,
|
| 1779 |
+
"learning_rate": 4.0020237794080445e-06,
|
| 1780 |
+
"loss": 3.027,
|
| 1781 |
+
"step": 6325
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.8031874525676701,
|
| 1785 |
+
"grad_norm": 1.664284110069275,
|
| 1786 |
+
"learning_rate": 3.9387806729066536e-06,
|
| 1787 |
+
"loss": 2.9782,
|
| 1788 |
+
"step": 6350
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.8063496078927397,
|
| 1792 |
+
"grad_norm": 1.7057468891143799,
|
| 1793 |
+
"learning_rate": 3.875537566405262e-06,
|
| 1794 |
+
"loss": 2.8451,
|
| 1795 |
+
"step": 6375
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.8095117632178093,
|
| 1799 |
+
"grad_norm": 1.548420786857605,
|
| 1800 |
+
"learning_rate": 3.812294459903871e-06,
|
| 1801 |
+
"loss": 2.7688,
|
| 1802 |
+
"step": 6400
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.8126739185428788,
|
| 1806 |
+
"grad_norm": 1.467391014099121,
|
| 1807 |
+
"learning_rate": 3.749051353402479e-06,
|
| 1808 |
+
"loss": 2.9415,
|
| 1809 |
+
"step": 6425
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.8158360738679484,
|
| 1813 |
+
"grad_norm": 1.6299670934677124,
|
| 1814 |
+
"learning_rate": 3.685808246901088e-06,
|
| 1815 |
+
"loss": 2.9104,
|
| 1816 |
+
"step": 6450
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.818998229193018,
|
| 1820 |
+
"grad_norm": 1.7904433012008667,
|
| 1821 |
+
"learning_rate": 3.622565140399697e-06,
|
| 1822 |
+
"loss": 2.8673,
|
| 1823 |
+
"step": 6475
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.8221603845180875,
|
| 1827 |
+
"grad_norm": 1.5487452745437622,
|
| 1828 |
+
"learning_rate": 3.5593220338983053e-06,
|
| 1829 |
+
"loss": 2.991,
|
| 1830 |
+
"step": 6500
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.8253225398431571,
|
| 1834 |
+
"grad_norm": 2.021498918533325,
|
| 1835 |
+
"learning_rate": 3.4960789273969143e-06,
|
| 1836 |
+
"loss": 2.9309,
|
| 1837 |
+
"step": 6525
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.8284846951682266,
|
| 1841 |
+
"grad_norm": 1.6797212362289429,
|
| 1842 |
+
"learning_rate": 3.4328358208955225e-06,
|
| 1843 |
+
"loss": 2.7787,
|
| 1844 |
+
"step": 6550
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.8316468504932962,
|
| 1848 |
+
"grad_norm": 1.5939486026763916,
|
| 1849 |
+
"learning_rate": 3.3695927143941316e-06,
|
| 1850 |
+
"loss": 2.8101,
|
| 1851 |
+
"step": 6575
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.8348090058183658,
|
| 1855 |
+
"grad_norm": 1.506430983543396,
|
| 1856 |
+
"learning_rate": 3.3063496078927398e-06,
|
| 1857 |
+
"loss": 2.9788,
|
| 1858 |
+
"step": 6600
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.8379711611434354,
|
| 1862 |
+
"grad_norm": 1.3282697200775146,
|
| 1863 |
+
"learning_rate": 3.243106501391349e-06,
|
| 1864 |
+
"loss": 2.8371,
|
| 1865 |
+
"step": 6625
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.841133316468505,
|
| 1869 |
+
"grad_norm": 1.799822449684143,
|
| 1870 |
+
"learning_rate": 3.179863394889957e-06,
|
| 1871 |
+
"loss": 2.9774,
|
| 1872 |
+
"step": 6650
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.8442954717935744,
|
| 1876 |
+
"grad_norm": 1.6544405221939087,
|
| 1877 |
+
"learning_rate": 3.116620288388566e-06,
|
| 1878 |
+
"loss": 2.8366,
|
| 1879 |
+
"step": 6675
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.847457627118644,
|
| 1883 |
+
"grad_norm": 1.7522649765014648,
|
| 1884 |
+
"learning_rate": 3.0533771818871742e-06,
|
| 1885 |
+
"loss": 2.9712,
|
| 1886 |
+
"step": 6700
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.8506197824437136,
|
| 1890 |
+
"grad_norm": 1.5311869382858276,
|
| 1891 |
+
"learning_rate": 2.9901340753857833e-06,
|
| 1892 |
+
"loss": 2.792,
|
| 1893 |
+
"step": 6725
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.8537819377687832,
|
| 1897 |
+
"grad_norm": 1.1723778247833252,
|
| 1898 |
+
"learning_rate": 2.9268909688843915e-06,
|
| 1899 |
+
"loss": 2.9395,
|
| 1900 |
+
"step": 6750
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.8569440930938528,
|
| 1904 |
+
"grad_norm": 1.9257378578186035,
|
| 1905 |
+
"learning_rate": 2.8636478623830005e-06,
|
| 1906 |
+
"loss": 2.9973,
|
| 1907 |
+
"step": 6775
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.8601062484189224,
|
| 1911 |
+
"grad_norm": 1.4320708513259888,
|
| 1912 |
+
"learning_rate": 2.8004047558816087e-06,
|
| 1913 |
+
"loss": 2.739,
|
| 1914 |
+
"step": 6800
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.8632684037439919,
|
| 1918 |
+
"grad_norm": 1.961350679397583,
|
| 1919 |
+
"learning_rate": 2.7371616493802178e-06,
|
| 1920 |
+
"loss": 3.0129,
|
| 1921 |
+
"step": 6825
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.8664305590690615,
|
| 1925 |
+
"grad_norm": 1.7785652875900269,
|
| 1926 |
+
"learning_rate": 2.673918542878827e-06,
|
| 1927 |
+
"loss": 3.0007,
|
| 1928 |
+
"step": 6850
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.869592714394131,
|
| 1932 |
+
"grad_norm": 1.5985599756240845,
|
| 1933 |
+
"learning_rate": 2.610675436377435e-06,
|
| 1934 |
+
"loss": 2.8996,
|
| 1935 |
+
"step": 6875
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.8727548697192006,
|
| 1939 |
+
"grad_norm": 1.4986529350280762,
|
| 1940 |
+
"learning_rate": 2.547432329876044e-06,
|
| 1941 |
+
"loss": 2.917,
|
| 1942 |
+
"step": 6900
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.8759170250442702,
|
| 1946 |
+
"grad_norm": 1.4311537742614746,
|
| 1947 |
+
"learning_rate": 2.4841892233746523e-06,
|
| 1948 |
+
"loss": 2.8844,
|
| 1949 |
+
"step": 6925
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.8790791803693397,
|
| 1953 |
+
"grad_norm": 1.8189743757247925,
|
| 1954 |
+
"learning_rate": 2.420946116873261e-06,
|
| 1955 |
+
"loss": 2.9588,
|
| 1956 |
+
"step": 6950
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.8822413356944093,
|
| 1960 |
+
"grad_norm": 1.1215569972991943,
|
| 1961 |
+
"learning_rate": 2.3577030103718695e-06,
|
| 1962 |
+
"loss": 2.8752,
|
| 1963 |
+
"step": 6975
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.8854034910194789,
|
| 1967 |
+
"grad_norm": 1.4951814413070679,
|
| 1968 |
+
"learning_rate": 2.294459903870478e-06,
|
| 1969 |
+
"loss": 2.8954,
|
| 1970 |
+
"step": 7000
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.8885656463445485,
|
| 1974 |
+
"grad_norm": 1.464539885520935,
|
| 1975 |
+
"learning_rate": 2.231216797369087e-06,
|
| 1976 |
+
"loss": 2.7321,
|
| 1977 |
+
"step": 7025
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.891727801669618,
|
| 1981 |
+
"grad_norm": 1.4621193408966064,
|
| 1982 |
+
"learning_rate": 2.1679736908676958e-06,
|
| 1983 |
+
"loss": 2.8688,
|
| 1984 |
+
"step": 7050
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.8948899569946875,
|
| 1988 |
+
"grad_norm": 1.6291460990905762,
|
| 1989 |
+
"learning_rate": 2.1047305843663044e-06,
|
| 1990 |
+
"loss": 2.8068,
|
| 1991 |
+
"step": 7075
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.8980521123197571,
|
| 1995 |
+
"grad_norm": 1.4216442108154297,
|
| 1996 |
+
"learning_rate": 2.041487477864913e-06,
|
| 1997 |
+
"loss": 2.9351,
|
| 1998 |
+
"step": 7100
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 0.9012142676448267,
|
| 2002 |
+
"grad_norm": 1.306211233139038,
|
| 2003 |
+
"learning_rate": 1.9782443713635216e-06,
|
| 2004 |
+
"loss": 2.9398,
|
| 2005 |
+
"step": 7125
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.9043764229698963,
|
| 2009 |
+
"grad_norm": 1.6560629606246948,
|
| 2010 |
+
"learning_rate": 1.9150012648621303e-06,
|
| 2011 |
+
"loss": 2.9728,
|
| 2012 |
+
"step": 7150
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.9075385782949659,
|
| 2016 |
+
"grad_norm": 1.9747501611709595,
|
| 2017 |
+
"learning_rate": 1.8517581583607389e-06,
|
| 2018 |
+
"loss": 2.903,
|
| 2019 |
+
"step": 7175
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 0.9107007336200355,
|
| 2023 |
+
"grad_norm": 1.3126964569091797,
|
| 2024 |
+
"learning_rate": 1.7885150518593475e-06,
|
| 2025 |
+
"loss": 2.796,
|
| 2026 |
+
"step": 7200
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.913862888945105,
|
| 2030 |
+
"grad_norm": 1.5026005506515503,
|
| 2031 |
+
"learning_rate": 1.7252719453579561e-06,
|
| 2032 |
+
"loss": 2.7565,
|
| 2033 |
+
"step": 7225
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 0.9170250442701745,
|
| 2037 |
+
"grad_norm": 1.5117318630218506,
|
| 2038 |
+
"learning_rate": 1.6620288388565647e-06,
|
| 2039 |
+
"loss": 2.8329,
|
| 2040 |
+
"step": 7250
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 0.9201871995952441,
|
| 2044 |
+
"grad_norm": 1.812044382095337,
|
| 2045 |
+
"learning_rate": 1.5987857323551734e-06,
|
| 2046 |
+
"loss": 2.9988,
|
| 2047 |
+
"step": 7275
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 0.9233493549203137,
|
| 2051 |
+
"grad_norm": 1.25447678565979,
|
| 2052 |
+
"learning_rate": 1.535542625853782e-06,
|
| 2053 |
+
"loss": 2.9395,
|
| 2054 |
+
"step": 7300
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 0.9265115102453833,
|
| 2058 |
+
"grad_norm": 1.6009807586669922,
|
| 2059 |
+
"learning_rate": 1.4722995193523906e-06,
|
| 2060 |
+
"loss": 2.8487,
|
| 2061 |
+
"step": 7325
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 0.9296736655704528,
|
| 2065 |
+
"grad_norm": 2.0983338356018066,
|
| 2066 |
+
"learning_rate": 1.4090564128509992e-06,
|
| 2067 |
+
"loss": 2.841,
|
| 2068 |
+
"step": 7350
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 0.9328358208955224,
|
| 2072 |
+
"grad_norm": 1.5755832195281982,
|
| 2073 |
+
"learning_rate": 1.3458133063496079e-06,
|
| 2074 |
+
"loss": 2.8646,
|
| 2075 |
+
"step": 7375
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 0.935997976220592,
|
| 2079 |
+
"grad_norm": 1.4814996719360352,
|
| 2080 |
+
"learning_rate": 1.2825701998482165e-06,
|
| 2081 |
+
"loss": 3.0229,
|
| 2082 |
+
"step": 7400
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.9391601315456615,
|
| 2086 |
+
"grad_norm": 1.687947392463684,
|
| 2087 |
+
"learning_rate": 1.2193270933468253e-06,
|
| 2088 |
+
"loss": 2.8668,
|
| 2089 |
+
"step": 7425
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 0.942322286870731,
|
| 2093 |
+
"grad_norm": 1.557085394859314,
|
| 2094 |
+
"learning_rate": 1.156083986845434e-06,
|
| 2095 |
+
"loss": 2.9006,
|
| 2096 |
+
"step": 7450
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.9454844421958006,
|
| 2100 |
+
"grad_norm": 1.4846750497817993,
|
| 2101 |
+
"learning_rate": 1.0928408803440425e-06,
|
| 2102 |
+
"loss": 2.9055,
|
| 2103 |
+
"step": 7475
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 0.9486465975208702,
|
| 2107 |
+
"grad_norm": 2.095038652420044,
|
| 2108 |
+
"learning_rate": 1.0295977738426512e-06,
|
| 2109 |
+
"loss": 3.0127,
|
| 2110 |
+
"step": 7500
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 0.9518087528459398,
|
| 2114 |
+
"grad_norm": 1.618682622909546,
|
| 2115 |
+
"learning_rate": 9.6635466734126e-07,
|
| 2116 |
+
"loss": 2.8352,
|
| 2117 |
+
"step": 7525
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 0.9549709081710094,
|
| 2121 |
+
"grad_norm": 1.6419885158538818,
|
| 2122 |
+
"learning_rate": 9.031115608398685e-07,
|
| 2123 |
+
"loss": 2.9209,
|
| 2124 |
+
"step": 7550
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 0.958133063496079,
|
| 2128 |
+
"grad_norm": 1.6612589359283447,
|
| 2129 |
+
"learning_rate": 8.398684543384771e-07,
|
| 2130 |
+
"loss": 2.8939,
|
| 2131 |
+
"step": 7575
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.9612952188211485,
|
| 2135 |
+
"grad_norm": 1.4516048431396484,
|
| 2136 |
+
"learning_rate": 7.766253478370859e-07,
|
| 2137 |
+
"loss": 2.9458,
|
| 2138 |
+
"step": 7600
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 0.964457374146218,
|
| 2142 |
+
"grad_norm": 1.864473581314087,
|
| 2143 |
+
"learning_rate": 7.133822413356945e-07,
|
| 2144 |
+
"loss": 2.958,
|
| 2145 |
+
"step": 7625
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 0.9676195294712876,
|
| 2149 |
+
"grad_norm": 1.5845332145690918,
|
| 2150 |
+
"learning_rate": 6.501391348343031e-07,
|
| 2151 |
+
"loss": 2.8868,
|
| 2152 |
+
"step": 7650
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 0.9707816847963572,
|
| 2156 |
+
"grad_norm": 1.4741908311843872,
|
| 2157 |
+
"learning_rate": 5.868960283329117e-07,
|
| 2158 |
+
"loss": 2.8258,
|
| 2159 |
+
"step": 7675
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 0.9739438401214268,
|
| 2163 |
+
"grad_norm": 1.5956405401229858,
|
| 2164 |
+
"learning_rate": 5.236529218315205e-07,
|
| 2165 |
+
"loss": 2.8274,
|
| 2166 |
+
"step": 7700
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 0.9771059954464963,
|
| 2170 |
+
"grad_norm": 1.5453468561172485,
|
| 2171 |
+
"learning_rate": 4.60409815330129e-07,
|
| 2172 |
+
"loss": 2.9071,
|
| 2173 |
+
"step": 7725
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 0.9802681507715659,
|
| 2177 |
+
"grad_norm": 1.745120882987976,
|
| 2178 |
+
"learning_rate": 3.9716670882873775e-07,
|
| 2179 |
+
"loss": 3.0668,
|
| 2180 |
+
"step": 7750
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 0.9834303060966355,
|
| 2184 |
+
"grad_norm": 1.3923741579055786,
|
| 2185 |
+
"learning_rate": 3.3392360232734637e-07,
|
| 2186 |
+
"loss": 3.0027,
|
| 2187 |
+
"step": 7775
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 0.986592461421705,
|
| 2191 |
+
"grad_norm": 1.4872705936431885,
|
| 2192 |
+
"learning_rate": 2.70680495825955e-07,
|
| 2193 |
+
"loss": 2.9386,
|
| 2194 |
+
"step": 7800
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 0.9897546167467746,
|
| 2198 |
+
"grad_norm": 2.8140368461608887,
|
| 2199 |
+
"learning_rate": 2.0743738932456364e-07,
|
| 2200 |
+
"loss": 2.7396,
|
| 2201 |
+
"step": 7825
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 0.9929167720718441,
|
| 2205 |
+
"grad_norm": 1.2857846021652222,
|
| 2206 |
+
"learning_rate": 1.4419428282317229e-07,
|
| 2207 |
+
"loss": 2.8295,
|
| 2208 |
+
"step": 7850
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 0.9960789273969137,
|
| 2212 |
+
"grad_norm": 1.593768835067749,
|
| 2213 |
+
"learning_rate": 8.095117632178092e-08,
|
| 2214 |
+
"loss": 3.0062,
|
| 2215 |
+
"step": 7875
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 0.9992410827219833,
|
| 2219 |
+
"grad_norm": 1.3941960334777832,
|
| 2220 |
+
"learning_rate": 1.770806982038958e-08,
|
| 2221 |
+
"loss": 2.7836,
|
| 2222 |
+
"step": 7900
|
| 2223 |
+
}
|
| 2224 |
+
],
|
| 2225 |
+
"logging_steps": 25,
|
| 2226 |
+
"max_steps": 7906,
|
| 2227 |
+
"num_input_tokens_seen": 0,
|
| 2228 |
+
"num_train_epochs": 1,
|
| 2229 |
+
"save_steps": 500,
|
| 2230 |
+
"stateful_callbacks": {
|
| 2231 |
+
"TrainerControl": {
|
| 2232 |
+
"args": {
|
| 2233 |
+
"should_epoch_stop": false,
|
| 2234 |
+
"should_evaluate": false,
|
| 2235 |
+
"should_log": false,
|
| 2236 |
+
"should_save": true,
|
| 2237 |
+
"should_training_stop": true
|
| 2238 |
+
},
|
| 2239 |
+
"attributes": {}
|
| 2240 |
+
}
|
| 2241 |
+
},
|
| 2242 |
+
"total_flos": 2103601100193792.0,
|
| 2243 |
+
"train_batch_size": 2,
|
| 2244 |
+
"trial_name": null,
|
| 2245 |
+
"trial_params": null
|
| 2246 |
+
}
|
checkpoint-7906/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:72b581cd8e50466800ca90569133129aafadaef26c6c1972c6abcf6f033d31ec
|
| 3 |
+
size 5869
|
config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_cross_attention": true,
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertLMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"gradient_checkpointing": false,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 384,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 1536,
|
| 15 |
+
"is_decoder": true,
|
| 16 |
+
"layer_norm_eps": 1e-12,
|
| 17 |
+
"max_position_embeddings": 512,
|
| 18 |
+
"model_type": "bert",
|
| 19 |
+
"num_attention_heads": 12,
|
| 20 |
+
"num_hidden_layers": 6,
|
| 21 |
+
"pad_token_id": 0,
|
| 22 |
+
"position_embedding_type": "absolute",
|
| 23 |
+
"transformers_version": "4.57.1",
|
| 24 |
+
"type_vocab_size": 2,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 30522
|
| 27 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"pad_token_id": 0,
|
| 4 |
+
"transformers_version": "4.57.1"
|
| 5 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a16e5f60ed44d8542a97a281d474481a19da2f9891decde86a56e18e6ae81b9
|
| 3 |
+
size 90990200
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": {
|
| 3 |
+
"content": "[CLS]",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"mask_token": {
|
| 10 |
+
"content": "[MASK]",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "[PAD]",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"sep_token": {
|
| 24 |
+
"content": "[SEP]",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"unk_token": {
|
| 31 |
+
"content": "[UNK]",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
}
|
| 37 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": false,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_basic_tokenize": true,
|
| 47 |
+
"do_lower_case": true,
|
| 48 |
+
"extra_special_tokens": {},
|
| 49 |
+
"mask_token": "[MASK]",
|
| 50 |
+
"max_length": 128,
|
| 51 |
+
"model_max_length": 256,
|
| 52 |
+
"never_split": null,
|
| 53 |
+
"pad_to_multiple_of": null,
|
| 54 |
+
"pad_token": "[PAD]",
|
| 55 |
+
"pad_token_type_id": 0,
|
| 56 |
+
"padding_side": "right",
|
| 57 |
+
"sep_token": "[SEP]",
|
| 58 |
+
"stride": 0,
|
| 59 |
+
"strip_accents": null,
|
| 60 |
+
"tokenize_chinese_chars": true,
|
| 61 |
+
"tokenizer_class": "BertTokenizer",
|
| 62 |
+
"truncation_side": "right",
|
| 63 |
+
"truncation_strategy": "longest_first",
|
| 64 |
+
"unk_token": "[UNK]"
|
| 65 |
+
}
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|