Training in progress, epoch 1, checkpoint
Browse files- last-checkpoint/config.json +31 -0
- last-checkpoint/generation_config.json +7 -0
- last-checkpoint/model.safetensors +3 -0
- last-checkpoint/optimizer.pt +3 -0
- last-checkpoint/rng_state.pth +3 -0
- last-checkpoint/scheduler.pt +3 -0
- last-checkpoint/trainer_state.json +1154 -0
- last-checkpoint/training_args.bin +3 -0
last-checkpoint/config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/opt-350m",
|
| 3 |
+
"_remove_final_layer_norm": false,
|
| 4 |
+
"activation_dropout": 0.0,
|
| 5 |
+
"activation_function": "relu",
|
| 6 |
+
"architectures": [
|
| 7 |
+
"OPTForCausalLM"
|
| 8 |
+
],
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"bos_token_id": 2,
|
| 11 |
+
"do_layer_norm_before": false,
|
| 12 |
+
"dropout": 0.1,
|
| 13 |
+
"enable_bias": true,
|
| 14 |
+
"eos_token_id": 2,
|
| 15 |
+
"ffn_dim": 4096,
|
| 16 |
+
"hidden_size": 1024,
|
| 17 |
+
"init_std": 0.02,
|
| 18 |
+
"layer_norm_elementwise_affine": true,
|
| 19 |
+
"layerdrop": 0.0,
|
| 20 |
+
"max_position_embeddings": 2048,
|
| 21 |
+
"model_type": "opt",
|
| 22 |
+
"num_attention_heads": 16,
|
| 23 |
+
"num_hidden_layers": 24,
|
| 24 |
+
"pad_token_id": 1,
|
| 25 |
+
"prefix": "</s>",
|
| 26 |
+
"torch_dtype": "float32",
|
| 27 |
+
"transformers_version": "4.47.1",
|
| 28 |
+
"use_cache": true,
|
| 29 |
+
"vocab_size": 50272,
|
| 30 |
+
"word_embed_proj_dim": 512
|
| 31 |
+
}
|
last-checkpoint/generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 2,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"pad_token_id": 1,
|
| 6 |
+
"transformers_version": "4.47.1"
|
| 7 |
+
}
|
last-checkpoint/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d1d7f3cfa105d55d981feea87dc1698d387783a26648d9c65ce26728eec859c4
|
| 3 |
+
size 1324830880
|
last-checkpoint/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:df180fbce5b0710b2f5ccb4689f53c435b2b47130f21918e3f901a58801e0b69
|
| 3 |
+
size 2649896094
|
last-checkpoint/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:97132a0bd0373c36b3bd93d6b28f9802a143f265a6eaa7b3097196f6002ba451
|
| 3 |
+
size 14244
|
last-checkpoint/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:11e3a63c9cfc5f5b20ed4deea6d33c221d465c076914b8f644e71e94dbcbcdbc
|
| 3 |
+
size 1064
|
last-checkpoint/trainer_state.json
ADDED
|
@@ -0,0 +1,1154 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.371297687292099,
|
| 3 |
+
"best_model_checkpoint": "./opt_trained/checkpoint-159",
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 159,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.006314127861089187,
|
| 13 |
+
"grad_norm": Infinity,
|
| 14 |
+
"learning_rate": 0.0001,
|
| 15 |
+
"loss": 6.0008,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.012628255722178374,
|
| 20 |
+
"grad_norm": 161.0778350830078,
|
| 21 |
+
"learning_rate": 9.984177215189874e-05,
|
| 22 |
+
"loss": 6.1896,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.018942383583267563,
|
| 27 |
+
"grad_norm": Infinity,
|
| 28 |
+
"learning_rate": 9.984177215189874e-05,
|
| 29 |
+
"loss": 20.1988,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.025256511444356748,
|
| 34 |
+
"grad_norm": Infinity,
|
| 35 |
+
"learning_rate": 9.984177215189874e-05,
|
| 36 |
+
"loss": 19.8097,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.03157063930544594,
|
| 41 |
+
"grad_norm": Infinity,
|
| 42 |
+
"learning_rate": 9.984177215189874e-05,
|
| 43 |
+
"loss": 19.9597,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.037884767166535126,
|
| 48 |
+
"grad_norm": 584.6585083007812,
|
| 49 |
+
"learning_rate": 9.968354430379747e-05,
|
| 50 |
+
"loss": 20.0818,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.04419889502762431,
|
| 55 |
+
"grad_norm": 723.80224609375,
|
| 56 |
+
"learning_rate": 9.952531645569621e-05,
|
| 57 |
+
"loss": 18.7822,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.050513022888713496,
|
| 62 |
+
"grad_norm": 608.1036376953125,
|
| 63 |
+
"learning_rate": 9.936708860759493e-05,
|
| 64 |
+
"loss": 27.5703,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.056827150749802685,
|
| 69 |
+
"grad_norm": 416.6692199707031,
|
| 70 |
+
"learning_rate": 9.920886075949367e-05,
|
| 71 |
+
"loss": 23.7453,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.06314127861089187,
|
| 76 |
+
"grad_norm": 377.489990234375,
|
| 77 |
+
"learning_rate": 9.90506329113924e-05,
|
| 78 |
+
"loss": 18.0302,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.06945540647198106,
|
| 83 |
+
"grad_norm": 187.46090698242188,
|
| 84 |
+
"learning_rate": 9.889240506329115e-05,
|
| 85 |
+
"loss": 13.8748,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.07576953433307025,
|
| 90 |
+
"grad_norm": 93.60346984863281,
|
| 91 |
+
"learning_rate": 9.873417721518988e-05,
|
| 92 |
+
"loss": 12.0784,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.08208366219415943,
|
| 97 |
+
"grad_norm": 99.17755889892578,
|
| 98 |
+
"learning_rate": 9.857594936708862e-05,
|
| 99 |
+
"loss": 10.016,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.08839779005524862,
|
| 104 |
+
"grad_norm": 191.79139709472656,
|
| 105 |
+
"learning_rate": 9.841772151898735e-05,
|
| 106 |
+
"loss": 9.5476,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.0947119179163378,
|
| 111 |
+
"grad_norm": 264.03436279296875,
|
| 112 |
+
"learning_rate": 9.825949367088608e-05,
|
| 113 |
+
"loss": 10.4265,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.10102604577742699,
|
| 118 |
+
"grad_norm": 212.4002685546875,
|
| 119 |
+
"learning_rate": 9.810126582278482e-05,
|
| 120 |
+
"loss": 10.2333,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.10734017363851618,
|
| 125 |
+
"grad_norm": 132.50941467285156,
|
| 126 |
+
"learning_rate": 9.794303797468355e-05,
|
| 127 |
+
"loss": 8.4413,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.11365430149960537,
|
| 132 |
+
"grad_norm": 197.4528350830078,
|
| 133 |
+
"learning_rate": 9.778481012658228e-05,
|
| 134 |
+
"loss": 8.8448,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.11996842936069456,
|
| 139 |
+
"grad_norm": 72.22266387939453,
|
| 140 |
+
"learning_rate": 9.762658227848101e-05,
|
| 141 |
+
"loss": 7.3288,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.12628255722178375,
|
| 146 |
+
"grad_norm": 99.05847930908203,
|
| 147 |
+
"learning_rate": 9.746835443037975e-05,
|
| 148 |
+
"loss": 6.6077,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.13259668508287292,
|
| 153 |
+
"grad_norm": 131.28546142578125,
|
| 154 |
+
"learning_rate": 9.731012658227849e-05,
|
| 155 |
+
"loss": 6.2343,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.13891081294396213,
|
| 160 |
+
"grad_norm": 38.381507873535156,
|
| 161 |
+
"learning_rate": 9.715189873417721e-05,
|
| 162 |
+
"loss": 4.9639,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.1452249408050513,
|
| 167 |
+
"grad_norm": 96.89019775390625,
|
| 168 |
+
"learning_rate": 9.699367088607595e-05,
|
| 169 |
+
"loss": 5.2425,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.1515390686661405,
|
| 174 |
+
"grad_norm": 91.1437759399414,
|
| 175 |
+
"learning_rate": 9.683544303797469e-05,
|
| 176 |
+
"loss": 4.7628,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.15785319652722968,
|
| 181 |
+
"grad_norm": 115.70115661621094,
|
| 182 |
+
"learning_rate": 9.667721518987343e-05,
|
| 183 |
+
"loss": 4.8574,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.16416732438831885,
|
| 188 |
+
"grad_norm": 54.2802734375,
|
| 189 |
+
"learning_rate": 9.651898734177216e-05,
|
| 190 |
+
"loss": 4.3247,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.17048145224940806,
|
| 195 |
+
"grad_norm": 83.99079895019531,
|
| 196 |
+
"learning_rate": 9.63607594936709e-05,
|
| 197 |
+
"loss": 4.6229,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.17679558011049723,
|
| 202 |
+
"grad_norm": 38.651798248291016,
|
| 203 |
+
"learning_rate": 9.620253164556962e-05,
|
| 204 |
+
"loss": 4.097,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.18310970797158643,
|
| 209 |
+
"grad_norm": 80.54218292236328,
|
| 210 |
+
"learning_rate": 9.604430379746836e-05,
|
| 211 |
+
"loss": 4.1475,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.1894238358326756,
|
| 216 |
+
"grad_norm": 57.01435470581055,
|
| 217 |
+
"learning_rate": 9.58860759493671e-05,
|
| 218 |
+
"loss": 4.1019,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.1957379636937648,
|
| 223 |
+
"grad_norm": 26.745500564575195,
|
| 224 |
+
"learning_rate": 9.572784810126582e-05,
|
| 225 |
+
"loss": 3.8805,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.20205209155485399,
|
| 230 |
+
"grad_norm": 31.093868255615234,
|
| 231 |
+
"learning_rate": 9.556962025316456e-05,
|
| 232 |
+
"loss": 3.8496,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.20836621941594316,
|
| 237 |
+
"grad_norm": 23.809965133666992,
|
| 238 |
+
"learning_rate": 9.541139240506329e-05,
|
| 239 |
+
"loss": 4.0042,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.21468034727703236,
|
| 244 |
+
"grad_norm": 21.90875816345215,
|
| 245 |
+
"learning_rate": 9.525316455696203e-05,
|
| 246 |
+
"loss": 3.4837,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.22099447513812154,
|
| 251 |
+
"grad_norm": 14.536842346191406,
|
| 252 |
+
"learning_rate": 9.509493670886075e-05,
|
| 253 |
+
"loss": 3.2669,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.22730860299921074,
|
| 258 |
+
"grad_norm": 19.5743408203125,
|
| 259 |
+
"learning_rate": 9.493670886075949e-05,
|
| 260 |
+
"loss": 3.5913,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.23362273086029992,
|
| 265 |
+
"grad_norm": 17.766374588012695,
|
| 266 |
+
"learning_rate": 9.477848101265823e-05,
|
| 267 |
+
"loss": 3.7308,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.23993685872138912,
|
| 272 |
+
"grad_norm": 17.416940689086914,
|
| 273 |
+
"learning_rate": 9.462025316455697e-05,
|
| 274 |
+
"loss": 3.5262,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.2462509865824783,
|
| 279 |
+
"grad_norm": 8.945268630981445,
|
| 280 |
+
"learning_rate": 9.446202531645571e-05,
|
| 281 |
+
"loss": 3.4766,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.2525651144435675,
|
| 286 |
+
"grad_norm": 10.85609245300293,
|
| 287 |
+
"learning_rate": 9.430379746835444e-05,
|
| 288 |
+
"loss": 3.4725,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.25887924230465664,
|
| 293 |
+
"grad_norm": 17.012737274169922,
|
| 294 |
+
"learning_rate": 9.414556962025317e-05,
|
| 295 |
+
"loss": 3.3638,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.26519337016574585,
|
| 300 |
+
"grad_norm": 9.961986541748047,
|
| 301 |
+
"learning_rate": 9.39873417721519e-05,
|
| 302 |
+
"loss": 3.4341,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.27150749802683505,
|
| 307 |
+
"grad_norm": 37.24739074707031,
|
| 308 |
+
"learning_rate": 9.382911392405064e-05,
|
| 309 |
+
"loss": 3.7622,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.27782162588792425,
|
| 314 |
+
"grad_norm": 40.56873321533203,
|
| 315 |
+
"learning_rate": 9.367088607594936e-05,
|
| 316 |
+
"loss": 3.4796,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.2841357537490134,
|
| 321 |
+
"grad_norm": 12.76440143585205,
|
| 322 |
+
"learning_rate": 9.35126582278481e-05,
|
| 323 |
+
"loss": 3.6249,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.2904498816101026,
|
| 328 |
+
"grad_norm": 27.08614730834961,
|
| 329 |
+
"learning_rate": 9.335443037974684e-05,
|
| 330 |
+
"loss": 3.5403,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.2967640094711918,
|
| 335 |
+
"grad_norm": 32.77756881713867,
|
| 336 |
+
"learning_rate": 9.319620253164557e-05,
|
| 337 |
+
"loss": 3.5529,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.303078137332281,
|
| 342 |
+
"grad_norm": 10.868654251098633,
|
| 343 |
+
"learning_rate": 9.303797468354431e-05,
|
| 344 |
+
"loss": 3.3178,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.30939226519337015,
|
| 349 |
+
"grad_norm": 28.319528579711914,
|
| 350 |
+
"learning_rate": 9.287974683544303e-05,
|
| 351 |
+
"loss": 3.3276,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.31570639305445936,
|
| 356 |
+
"grad_norm": 27.239425659179688,
|
| 357 |
+
"learning_rate": 9.272151898734177e-05,
|
| 358 |
+
"loss": 3.5792,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.32202052091554856,
|
| 363 |
+
"grad_norm": 7.753026962280273,
|
| 364 |
+
"learning_rate": 9.256329113924051e-05,
|
| 365 |
+
"loss": 3.5363,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.3283346487766377,
|
| 370 |
+
"grad_norm": 18.737960815429688,
|
| 371 |
+
"learning_rate": 9.240506329113925e-05,
|
| 372 |
+
"loss": 3.3422,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.3346487766377269,
|
| 377 |
+
"grad_norm": 14.71604061126709,
|
| 378 |
+
"learning_rate": 9.224683544303798e-05,
|
| 379 |
+
"loss": 3.2696,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.3409629044988161,
|
| 384 |
+
"grad_norm": 15.523343086242676,
|
| 385 |
+
"learning_rate": 9.208860759493671e-05,
|
| 386 |
+
"loss": 3.5421,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.3472770323599053,
|
| 391 |
+
"grad_norm": 11.16079044342041,
|
| 392 |
+
"learning_rate": 9.193037974683545e-05,
|
| 393 |
+
"loss": 3.3622,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.35359116022099446,
|
| 398 |
+
"grad_norm": 9.385275840759277,
|
| 399 |
+
"learning_rate": 9.177215189873418e-05,
|
| 400 |
+
"loss": 3.3368,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.35990528808208366,
|
| 405 |
+
"grad_norm": 7.030306816101074,
|
| 406 |
+
"learning_rate": 9.161392405063292e-05,
|
| 407 |
+
"loss": 3.383,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.36621941594317287,
|
| 412 |
+
"grad_norm": 8.905327796936035,
|
| 413 |
+
"learning_rate": 9.145569620253164e-05,
|
| 414 |
+
"loss": 3.4087,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.372533543804262,
|
| 419 |
+
"grad_norm": 13.373973846435547,
|
| 420 |
+
"learning_rate": 9.129746835443038e-05,
|
| 421 |
+
"loss": 3.407,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.3788476716653512,
|
| 426 |
+
"grad_norm": 7.750667095184326,
|
| 427 |
+
"learning_rate": 9.113924050632912e-05,
|
| 428 |
+
"loss": 3.3293,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.3851617995264404,
|
| 433 |
+
"grad_norm": 23.99922752380371,
|
| 434 |
+
"learning_rate": 9.098101265822785e-05,
|
| 435 |
+
"loss": 3.385,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.3914759273875296,
|
| 440 |
+
"grad_norm": 20.64397430419922,
|
| 441 |
+
"learning_rate": 9.082278481012659e-05,
|
| 442 |
+
"loss": 3.4648,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.39779005524861877,
|
| 447 |
+
"grad_norm": 8.52364444732666,
|
| 448 |
+
"learning_rate": 9.066455696202531e-05,
|
| 449 |
+
"loss": 3.3165,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.40410418310970797,
|
| 454 |
+
"grad_norm": 10.17910385131836,
|
| 455 |
+
"learning_rate": 9.050632911392407e-05,
|
| 456 |
+
"loss": 3.3308,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.4104183109707972,
|
| 461 |
+
"grad_norm": 11.811700820922852,
|
| 462 |
+
"learning_rate": 9.034810126582279e-05,
|
| 463 |
+
"loss": 3.316,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.4167324388318863,
|
| 468 |
+
"grad_norm": 6.778939723968506,
|
| 469 |
+
"learning_rate": 9.018987341772153e-05,
|
| 470 |
+
"loss": 3.0975,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.4230465666929755,
|
| 475 |
+
"grad_norm": 5.862908840179443,
|
| 476 |
+
"learning_rate": 9.003164556962026e-05,
|
| 477 |
+
"loss": 3.1885,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.4293606945540647,
|
| 482 |
+
"grad_norm": 5.790811538696289,
|
| 483 |
+
"learning_rate": 8.9873417721519e-05,
|
| 484 |
+
"loss": 3.2072,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.43567482241515393,
|
| 489 |
+
"grad_norm": 5.9749064445495605,
|
| 490 |
+
"learning_rate": 8.971518987341772e-05,
|
| 491 |
+
"loss": 3.0589,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.4419889502762431,
|
| 496 |
+
"grad_norm": 11.786849975585938,
|
| 497 |
+
"learning_rate": 8.955696202531646e-05,
|
| 498 |
+
"loss": 3.188,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.4483030781373323,
|
| 503 |
+
"grad_norm": 15.524914741516113,
|
| 504 |
+
"learning_rate": 8.93987341772152e-05,
|
| 505 |
+
"loss": 3.2465,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.4546172059984215,
|
| 510 |
+
"grad_norm": 14.146249771118164,
|
| 511 |
+
"learning_rate": 8.924050632911392e-05,
|
| 512 |
+
"loss": 3.2995,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.46093133385951063,
|
| 517 |
+
"grad_norm": 3.7766897678375244,
|
| 518 |
+
"learning_rate": 8.908227848101266e-05,
|
| 519 |
+
"loss": 3.2287,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.46724546172059983,
|
| 524 |
+
"grad_norm": 13.466503143310547,
|
| 525 |
+
"learning_rate": 8.892405063291139e-05,
|
| 526 |
+
"loss": 3.2794,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.47355958958168903,
|
| 531 |
+
"grad_norm": 9.60093879699707,
|
| 532 |
+
"learning_rate": 8.876582278481013e-05,
|
| 533 |
+
"loss": 3.2478,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.47987371744277824,
|
| 538 |
+
"grad_norm": 8.259520530700684,
|
| 539 |
+
"learning_rate": 8.860759493670887e-05,
|
| 540 |
+
"loss": 3.3586,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.4861878453038674,
|
| 545 |
+
"grad_norm": 6.9965410232543945,
|
| 546 |
+
"learning_rate": 8.84493670886076e-05,
|
| 547 |
+
"loss": 3.2365,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.4925019731649566,
|
| 552 |
+
"grad_norm": 12.788185119628906,
|
| 553 |
+
"learning_rate": 8.829113924050633e-05,
|
| 554 |
+
"loss": 3.2107,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.4988161010260458,
|
| 559 |
+
"grad_norm": 10.677863121032715,
|
| 560 |
+
"learning_rate": 8.813291139240507e-05,
|
| 561 |
+
"loss": 3.2843,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.505130228887135,
|
| 566 |
+
"grad_norm": 8.940823554992676,
|
| 567 |
+
"learning_rate": 8.797468354430381e-05,
|
| 568 |
+
"loss": 3.1655,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.5114443567482242,
|
| 573 |
+
"grad_norm": 4.832064628601074,
|
| 574 |
+
"learning_rate": 8.781645569620253e-05,
|
| 575 |
+
"loss": 3.1706,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.5177584846093133,
|
| 580 |
+
"grad_norm": 10.519792556762695,
|
| 581 |
+
"learning_rate": 8.765822784810127e-05,
|
| 582 |
+
"loss": 3.2113,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.5240726124704025,
|
| 587 |
+
"grad_norm": 21.498655319213867,
|
| 588 |
+
"learning_rate": 8.75e-05,
|
| 589 |
+
"loss": 3.0548,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.5303867403314917,
|
| 594 |
+
"grad_norm": 16.82743263244629,
|
| 595 |
+
"learning_rate": 8.734177215189874e-05,
|
| 596 |
+
"loss": 3.0727,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.5367008681925809,
|
| 601 |
+
"grad_norm": 17.08803367614746,
|
| 602 |
+
"learning_rate": 8.718354430379748e-05,
|
| 603 |
+
"loss": 3.1257,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.5430149960536701,
|
| 608 |
+
"grad_norm": 8.622400283813477,
|
| 609 |
+
"learning_rate": 8.70253164556962e-05,
|
| 610 |
+
"loss": 3.1575,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.5493291239147593,
|
| 615 |
+
"grad_norm": 17.043874740600586,
|
| 616 |
+
"learning_rate": 8.686708860759494e-05,
|
| 617 |
+
"loss": 3.2683,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.5556432517758485,
|
| 622 |
+
"grad_norm": 11.532078742980957,
|
| 623 |
+
"learning_rate": 8.670886075949367e-05,
|
| 624 |
+
"loss": 3.1915,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.5619573796369376,
|
| 629 |
+
"grad_norm": 23.534517288208008,
|
| 630 |
+
"learning_rate": 8.65506329113924e-05,
|
| 631 |
+
"loss": 3.2199,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.5682715074980268,
|
| 636 |
+
"grad_norm": 20.578760147094727,
|
| 637 |
+
"learning_rate": 8.639240506329115e-05,
|
| 638 |
+
"loss": 3.052,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.574585635359116,
|
| 643 |
+
"grad_norm": 7.6293535232543945,
|
| 644 |
+
"learning_rate": 8.623417721518988e-05,
|
| 645 |
+
"loss": 3.1255,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.5808997632202052,
|
| 650 |
+
"grad_norm": 13.604530334472656,
|
| 651 |
+
"learning_rate": 8.607594936708861e-05,
|
| 652 |
+
"loss": 3.2718,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.5872138910812944,
|
| 657 |
+
"grad_norm": 9.306023597717285,
|
| 658 |
+
"learning_rate": 8.591772151898735e-05,
|
| 659 |
+
"loss": 3.2007,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.5935280189423836,
|
| 664 |
+
"grad_norm": 15.916861534118652,
|
| 665 |
+
"learning_rate": 8.575949367088609e-05,
|
| 666 |
+
"loss": 3.0144,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.5998421468034728,
|
| 671 |
+
"grad_norm": 11.088306427001953,
|
| 672 |
+
"learning_rate": 8.560126582278481e-05,
|
| 673 |
+
"loss": 3.0094,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.606156274664562,
|
| 678 |
+
"grad_norm": 12.851061820983887,
|
| 679 |
+
"learning_rate": 8.544303797468355e-05,
|
| 680 |
+
"loss": 3.0338,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.6124704025256511,
|
| 685 |
+
"grad_norm": 8.163888931274414,
|
| 686 |
+
"learning_rate": 8.528481012658228e-05,
|
| 687 |
+
"loss": 3.0715,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.6187845303867403,
|
| 692 |
+
"grad_norm": 10.81649398803711,
|
| 693 |
+
"learning_rate": 8.512658227848102e-05,
|
| 694 |
+
"loss": 3.1136,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.6250986582478295,
|
| 699 |
+
"grad_norm": 4.056423664093018,
|
| 700 |
+
"learning_rate": 8.496835443037974e-05,
|
| 701 |
+
"loss": 3.1062,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.6314127861089187,
|
| 706 |
+
"grad_norm": 10.310821533203125,
|
| 707 |
+
"learning_rate": 8.481012658227848e-05,
|
| 708 |
+
"loss": 3.1323,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.6377269139700079,
|
| 713 |
+
"grad_norm": 4.946181774139404,
|
| 714 |
+
"learning_rate": 8.465189873417722e-05,
|
| 715 |
+
"loss": 3.112,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.6440410418310971,
|
| 720 |
+
"grad_norm": 12.03683853149414,
|
| 721 |
+
"learning_rate": 8.449367088607595e-05,
|
| 722 |
+
"loss": 3.0755,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.6503551696921863,
|
| 727 |
+
"grad_norm": 8.58892822265625,
|
| 728 |
+
"learning_rate": 8.43354430379747e-05,
|
| 729 |
+
"loss": 3.114,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.6566692975532754,
|
| 734 |
+
"grad_norm": 4.111664295196533,
|
| 735 |
+
"learning_rate": 8.417721518987342e-05,
|
| 736 |
+
"loss": 3.1824,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.6629834254143646,
|
| 741 |
+
"grad_norm": 8.031673431396484,
|
| 742 |
+
"learning_rate": 8.401898734177216e-05,
|
| 743 |
+
"loss": 3.0504,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.6692975532754538,
|
| 748 |
+
"grad_norm": 8.085382461547852,
|
| 749 |
+
"learning_rate": 8.386075949367089e-05,
|
| 750 |
+
"loss": 2.9141,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.675611681136543,
|
| 755 |
+
"grad_norm": 3.866814374923706,
|
| 756 |
+
"learning_rate": 8.370253164556963e-05,
|
| 757 |
+
"loss": 2.9354,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.6819258089976322,
|
| 762 |
+
"grad_norm": 21.265792846679688,
|
| 763 |
+
"learning_rate": 8.354430379746835e-05,
|
| 764 |
+
"loss": 3.0994,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.6882399368587214,
|
| 769 |
+
"grad_norm": 11.916318893432617,
|
| 770 |
+
"learning_rate": 8.33860759493671e-05,
|
| 771 |
+
"loss": 3.1088,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.6945540647198106,
|
| 776 |
+
"grad_norm": 21.871856689453125,
|
| 777 |
+
"learning_rate": 8.322784810126583e-05,
|
| 778 |
+
"loss": 3.1318,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.7008681925808997,
|
| 783 |
+
"grad_norm": 6.908069610595703,
|
| 784 |
+
"learning_rate": 8.306962025316456e-05,
|
| 785 |
+
"loss": 3.1491,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.7071823204419889,
|
| 790 |
+
"grad_norm": 36.063926696777344,
|
| 791 |
+
"learning_rate": 8.29113924050633e-05,
|
| 792 |
+
"loss": 3.4333,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.7134964483030781,
|
| 797 |
+
"grad_norm": 20.55953025817871,
|
| 798 |
+
"learning_rate": 8.275316455696202e-05,
|
| 799 |
+
"loss": 3.2292,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.7198105761641673,
|
| 804 |
+
"grad_norm": 15.739864349365234,
|
| 805 |
+
"learning_rate": 8.259493670886076e-05,
|
| 806 |
+
"loss": 2.993,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.7261247040252565,
|
| 811 |
+
"grad_norm": 15.668102264404297,
|
| 812 |
+
"learning_rate": 8.243670886075949e-05,
|
| 813 |
+
"loss": 2.9661,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.7324388318863457,
|
| 818 |
+
"grad_norm": 13.283061981201172,
|
| 819 |
+
"learning_rate": 8.227848101265824e-05,
|
| 820 |
+
"loss": 3.3498,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.7387529597474349,
|
| 825 |
+
"grad_norm": 10.03203296661377,
|
| 826 |
+
"learning_rate": 8.212025316455697e-05,
|
| 827 |
+
"loss": 3.0689,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.745067087608524,
|
| 832 |
+
"grad_norm": 7.466861248016357,
|
| 833 |
+
"learning_rate": 8.19620253164557e-05,
|
| 834 |
+
"loss": 3.0641,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.7513812154696132,
|
| 839 |
+
"grad_norm": 12.9628267288208,
|
| 840 |
+
"learning_rate": 8.180379746835444e-05,
|
| 841 |
+
"loss": 2.982,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.7576953433307024,
|
| 846 |
+
"grad_norm": 6.9294023513793945,
|
| 847 |
+
"learning_rate": 8.164556962025317e-05,
|
| 848 |
+
"loss": 3.1472,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.7640094711917916,
|
| 853 |
+
"grad_norm": 13.094769477844238,
|
| 854 |
+
"learning_rate": 8.148734177215191e-05,
|
| 855 |
+
"loss": 3.1785,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.7703235990528808,
|
| 860 |
+
"grad_norm": 15.382439613342285,
|
| 861 |
+
"learning_rate": 8.132911392405063e-05,
|
| 862 |
+
"loss": 3.1517,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.77663772691397,
|
| 867 |
+
"grad_norm": 3.922220230102539,
|
| 868 |
+
"learning_rate": 8.117088607594937e-05,
|
| 869 |
+
"loss": 3.0825,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.7829518547750592,
|
| 874 |
+
"grad_norm": 13.367452621459961,
|
| 875 |
+
"learning_rate": 8.10126582278481e-05,
|
| 876 |
+
"loss": 3.1165,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.7892659826361483,
|
| 881 |
+
"grad_norm": 13.817737579345703,
|
| 882 |
+
"learning_rate": 8.085443037974684e-05,
|
| 883 |
+
"loss": 3.3522,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.7955801104972375,
|
| 888 |
+
"grad_norm": 6.693469524383545,
|
| 889 |
+
"learning_rate": 8.069620253164558e-05,
|
| 890 |
+
"loss": 2.9785,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.8018942383583267,
|
| 895 |
+
"grad_norm": 6.970970630645752,
|
| 896 |
+
"learning_rate": 8.05379746835443e-05,
|
| 897 |
+
"loss": 3.2253,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.8082083662194159,
|
| 902 |
+
"grad_norm": 6.731435775756836,
|
| 903 |
+
"learning_rate": 8.037974683544304e-05,
|
| 904 |
+
"loss": 3.323,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.8145224940805051,
|
| 909 |
+
"grad_norm": 4.875186920166016,
|
| 910 |
+
"learning_rate": 8.022151898734177e-05,
|
| 911 |
+
"loss": 3.1744,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.8208366219415943,
|
| 916 |
+
"grad_norm": 9.212851524353027,
|
| 917 |
+
"learning_rate": 8.006329113924052e-05,
|
| 918 |
+
"loss": 3.3191,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.8271507498026835,
|
| 923 |
+
"grad_norm": 4.607917785644531,
|
| 924 |
+
"learning_rate": 7.990506329113924e-05,
|
| 925 |
+
"loss": 3.1394,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.8334648776637726,
|
| 930 |
+
"grad_norm": 14.266739845275879,
|
| 931 |
+
"learning_rate": 7.974683544303798e-05,
|
| 932 |
+
"loss": 2.985,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.8397790055248618,
|
| 937 |
+
"grad_norm": 7.989418983459473,
|
| 938 |
+
"learning_rate": 7.958860759493671e-05,
|
| 939 |
+
"loss": 2.882,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.846093133385951,
|
| 944 |
+
"grad_norm": 7.034262657165527,
|
| 945 |
+
"learning_rate": 7.943037974683545e-05,
|
| 946 |
+
"loss": 3.1717,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.8524072612470402,
|
| 951 |
+
"grad_norm": 7.286618709564209,
|
| 952 |
+
"learning_rate": 7.927215189873419e-05,
|
| 953 |
+
"loss": 3.0567,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.8587213891081295,
|
| 958 |
+
"grad_norm": 4.958951473236084,
|
| 959 |
+
"learning_rate": 7.911392405063291e-05,
|
| 960 |
+
"loss": 3.0012,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.8650355169692187,
|
| 965 |
+
"grad_norm": 2.9031686782836914,
|
| 966 |
+
"learning_rate": 7.895569620253165e-05,
|
| 967 |
+
"loss": 3.0797,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.8713496448303079,
|
| 972 |
+
"grad_norm": 6.901837348937988,
|
| 973 |
+
"learning_rate": 7.879746835443038e-05,
|
| 974 |
+
"loss": 3.2493,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.877663772691397,
|
| 979 |
+
"grad_norm": 3.709942579269409,
|
| 980 |
+
"learning_rate": 7.863924050632912e-05,
|
| 981 |
+
"loss": 3.165,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.8839779005524862,
|
| 986 |
+
"grad_norm": 7.831309795379639,
|
| 987 |
+
"learning_rate": 7.848101265822784e-05,
|
| 988 |
+
"loss": 3.1126,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.8902920284135754,
|
| 993 |
+
"grad_norm": 8.239750862121582,
|
| 994 |
+
"learning_rate": 7.832278481012658e-05,
|
| 995 |
+
"loss": 2.9461,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.8966061562746646,
|
| 1000 |
+
"grad_norm": 5.883851528167725,
|
| 1001 |
+
"learning_rate": 7.816455696202532e-05,
|
| 1002 |
+
"loss": 3.1751,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.9029202841357538,
|
| 1007 |
+
"grad_norm": 4.314850330352783,
|
| 1008 |
+
"learning_rate": 7.800632911392406e-05,
|
| 1009 |
+
"loss": 3.1033,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.909234411996843,
|
| 1014 |
+
"grad_norm": 4.290492534637451,
|
| 1015 |
+
"learning_rate": 7.78481012658228e-05,
|
| 1016 |
+
"loss": 2.9642,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.9155485398579322,
|
| 1021 |
+
"grad_norm": 5.874259948730469,
|
| 1022 |
+
"learning_rate": 7.768987341772152e-05,
|
| 1023 |
+
"loss": 3.0721,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.9218626677190213,
|
| 1028 |
+
"grad_norm": 3.419154644012451,
|
| 1029 |
+
"learning_rate": 7.753164556962026e-05,
|
| 1030 |
+
"loss": 2.8922,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.9281767955801105,
|
| 1035 |
+
"grad_norm": 7.392196178436279,
|
| 1036 |
+
"learning_rate": 7.737341772151899e-05,
|
| 1037 |
+
"loss": 3.3249,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.9344909234411997,
|
| 1042 |
+
"grad_norm": 8.384613990783691,
|
| 1043 |
+
"learning_rate": 7.721518987341773e-05,
|
| 1044 |
+
"loss": 3.177,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.9408050513022889,
|
| 1049 |
+
"grad_norm": 3.6932125091552734,
|
| 1050 |
+
"learning_rate": 7.705696202531645e-05,
|
| 1051 |
+
"loss": 2.9024,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.9471191791633781,
|
| 1056 |
+
"grad_norm": 10.121501922607422,
|
| 1057 |
+
"learning_rate": 7.689873417721519e-05,
|
| 1058 |
+
"loss": 2.8811,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.9534333070244673,
|
| 1063 |
+
"grad_norm": 6.764036655426025,
|
| 1064 |
+
"learning_rate": 7.674050632911393e-05,
|
| 1065 |
+
"loss": 3.0374,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.9597474348855565,
|
| 1070 |
+
"grad_norm": 8.226846694946289,
|
| 1071 |
+
"learning_rate": 7.658227848101266e-05,
|
| 1072 |
+
"loss": 3.1455,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.9660615627466457,
|
| 1077 |
+
"grad_norm": 7.442160129547119,
|
| 1078 |
+
"learning_rate": 7.64240506329114e-05,
|
| 1079 |
+
"loss": 3.2016,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.9723756906077348,
|
| 1084 |
+
"grad_norm": 4.500311851501465,
|
| 1085 |
+
"learning_rate": 7.626582278481012e-05,
|
| 1086 |
+
"loss": 3.1067,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.978689818468824,
|
| 1091 |
+
"grad_norm": 4.05573034286499,
|
| 1092 |
+
"learning_rate": 7.610759493670886e-05,
|
| 1093 |
+
"loss": 3.0543,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.9850039463299132,
|
| 1098 |
+
"grad_norm": 5.813141345977783,
|
| 1099 |
+
"learning_rate": 7.59493670886076e-05,
|
| 1100 |
+
"loss": 3.0306,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.9913180741910024,
|
| 1105 |
+
"grad_norm": 9.513184547424316,
|
| 1106 |
+
"learning_rate": 7.579113924050634e-05,
|
| 1107 |
+
"loss": 2.8891,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.9976322020520916,
|
| 1112 |
+
"grad_norm": 4.737900257110596,
|
| 1113 |
+
"learning_rate": 7.563291139240506e-05,
|
| 1114 |
+
"loss": 3.142,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 1.0,
|
| 1119 |
+
"grad_norm": 1.998995304107666,
|
| 1120 |
+
"learning_rate": 7.54746835443038e-05,
|
| 1121 |
+
"loss": 1.1642,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 1.0,
|
| 1126 |
+
"eval_loss": 0.371297687292099,
|
| 1127 |
+
"eval_runtime": 19.3489,
|
| 1128 |
+
"eval_samples_per_second": 20.673,
|
| 1129 |
+
"eval_steps_per_second": 5.168,
|
| 1130 |
+
"step": 159
|
| 1131 |
+
}
|
| 1132 |
+
],
|
| 1133 |
+
"logging_steps": 1,
|
| 1134 |
+
"max_steps": 632,
|
| 1135 |
+
"num_input_tokens_seen": 0,
|
| 1136 |
+
"num_train_epochs": 4,
|
| 1137 |
+
"save_steps": 500,
|
| 1138 |
+
"stateful_callbacks": {
|
| 1139 |
+
"TrainerControl": {
|
| 1140 |
+
"args": {
|
| 1141 |
+
"should_epoch_stop": false,
|
| 1142 |
+
"should_evaluate": false,
|
| 1143 |
+
"should_log": false,
|
| 1144 |
+
"should_save": true,
|
| 1145 |
+
"should_training_stop": false
|
| 1146 |
+
},
|
| 1147 |
+
"attributes": {}
|
| 1148 |
+
}
|
| 1149 |
+
},
|
| 1150 |
+
"total_flos": 1.0705428106739712e+16,
|
| 1151 |
+
"train_batch_size": 6,
|
| 1152 |
+
"trial_name": null,
|
| 1153 |
+
"trial_params": null
|
| 1154 |
+
}
|
last-checkpoint/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0ab77a5eae38b0e271c42e2acd0600243694639d1424fbedb4fd0a49e94c28e2
|
| 3 |
+
size 5368
|