Push getvar generic t5 model
Browse files- json_extraction_point_activity/checkpoint-1000/config.json +61 -0
- json_extraction_point_activity/checkpoint-1000/generation_config.json +7 -0
- json_extraction_point_activity/checkpoint-1000/model.safetensors +3 -0
- json_extraction_point_activity/checkpoint-1000/optimizer.pt +3 -0
- json_extraction_point_activity/checkpoint-1000/rng_state.pth +3 -0
- json_extraction_point_activity/checkpoint-1000/scheduler.pt +3 -0
- json_extraction_point_activity/checkpoint-1000/trainer_state.json +1005 -0
- json_extraction_point_activity/checkpoint-1000/training_args.bin +3 -0
- json_extraction_point_activity/checkpoint-1015/config.json +61 -0
- json_extraction_point_activity/checkpoint-1015/generation_config.json +7 -0
- json_extraction_point_activity/checkpoint-1015/model.safetensors +3 -0
- json_extraction_point_activity/checkpoint-1015/optimizer.pt +3 -0
- json_extraction_point_activity/checkpoint-1015/rng_state.pth +3 -0
- json_extraction_point_activity/checkpoint-1015/scheduler.pt +3 -0
- json_extraction_point_activity/checkpoint-1015/trainer_state.json +1012 -0
- json_extraction_point_activity/checkpoint-1015/training_args.bin +3 -0
- json_extraction_point_activity/checkpoint-500/model.safetensors +1 -1
- json_extraction_point_activity/checkpoint-500/optimizer.pt +1 -1
- json_extraction_point_activity/checkpoint-500/rng_state.pth +1 -1
- json_extraction_point_activity/checkpoint-500/scheduler.pt +1 -1
- json_extraction_point_activity/checkpoint-500/trainer_state.json +312 -336
- json_extraction_point_activity/model.safetensors +1 -1
json_extraction_point_activity/checkpoint-1000/config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "t5-large",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"T5ForConditionalGeneration"
|
| 5 |
+
],
|
| 6 |
+
"classifier_dropout": 0.0,
|
| 7 |
+
"d_ff": 4096,
|
| 8 |
+
"d_kv": 64,
|
| 9 |
+
"d_model": 1024,
|
| 10 |
+
"decoder_start_token_id": 0,
|
| 11 |
+
"dense_act_fn": "relu",
|
| 12 |
+
"dropout_rate": 0.1,
|
| 13 |
+
"eos_token_id": 1,
|
| 14 |
+
"feed_forward_proj": "relu",
|
| 15 |
+
"initializer_factor": 1.0,
|
| 16 |
+
"is_encoder_decoder": true,
|
| 17 |
+
"is_gated_act": false,
|
| 18 |
+
"layer_norm_epsilon": 1e-06,
|
| 19 |
+
"model_type": "t5",
|
| 20 |
+
"n_positions": 512,
|
| 21 |
+
"num_decoder_layers": 24,
|
| 22 |
+
"num_heads": 16,
|
| 23 |
+
"num_layers": 24,
|
| 24 |
+
"output_past": true,
|
| 25 |
+
"pad_token_id": 0,
|
| 26 |
+
"relative_attention_max_distance": 128,
|
| 27 |
+
"relative_attention_num_buckets": 32,
|
| 28 |
+
"task_specific_params": {
|
| 29 |
+
"summarization": {
|
| 30 |
+
"early_stopping": true,
|
| 31 |
+
"length_penalty": 2.0,
|
| 32 |
+
"max_length": 200,
|
| 33 |
+
"min_length": 30,
|
| 34 |
+
"no_repeat_ngram_size": 3,
|
| 35 |
+
"num_beams": 4,
|
| 36 |
+
"prefix": "summarize: "
|
| 37 |
+
},
|
| 38 |
+
"translation_en_to_de": {
|
| 39 |
+
"early_stopping": true,
|
| 40 |
+
"max_length": 300,
|
| 41 |
+
"num_beams": 4,
|
| 42 |
+
"prefix": "translate English to German: "
|
| 43 |
+
},
|
| 44 |
+
"translation_en_to_fr": {
|
| 45 |
+
"early_stopping": true,
|
| 46 |
+
"max_length": 300,
|
| 47 |
+
"num_beams": 4,
|
| 48 |
+
"prefix": "translate English to French: "
|
| 49 |
+
},
|
| 50 |
+
"translation_en_to_ro": {
|
| 51 |
+
"early_stopping": true,
|
| 52 |
+
"max_length": 300,
|
| 53 |
+
"num_beams": 4,
|
| 54 |
+
"prefix": "translate English to Romanian: "
|
| 55 |
+
}
|
| 56 |
+
},
|
| 57 |
+
"torch_dtype": "float32",
|
| 58 |
+
"transformers_version": "4.48.2",
|
| 59 |
+
"use_cache": true,
|
| 60 |
+
"vocab_size": 32128
|
| 61 |
+
}
|
json_extraction_point_activity/checkpoint-1000/generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"decoder_start_token_id": 0,
|
| 4 |
+
"eos_token_id": 1,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.48.2"
|
| 7 |
+
}
|
json_extraction_point_activity/checkpoint-1000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f2033e1d91207701dcf2a50652e20805abce211fa98d7df187b671b1ef7e7783
|
| 3 |
+
size 2950734544
|
json_extraction_point_activity/checkpoint-1000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:304f67deadb2dd9d193f2e60ec9ab4c75474742551d5a0b0cdf99c290e5c949d
|
| 3 |
+
size 5901778825
|
json_extraction_point_activity/checkpoint-1000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5a17102bba356f793b2331a99300071e6d313d5a727155ed5f9df148159ac27b
|
| 3 |
+
size 14244
|
json_extraction_point_activity/checkpoint-1000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5f68200a6cbb8b4b695a170a8030cf83701b3b906f43a823586949eff641b80a
|
| 3 |
+
size 1064
|
json_extraction_point_activity/checkpoint-1000/trainer_state.json
ADDED
|
@@ -0,0 +1,1005 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 34.48275862068966,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.3448275862068966,
|
| 13 |
+
"grad_norm": 174.2040557861328,
|
| 14 |
+
"learning_rate": 1.9802955665024632e-05,
|
| 15 |
+
"loss": 11.9232,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.6896551724137931,
|
| 20 |
+
"grad_norm": 54.92245101928711,
|
| 21 |
+
"learning_rate": 1.9605911330049263e-05,
|
| 22 |
+
"loss": 6.0635,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 1.0,
|
| 27 |
+
"eval_loss": 0.8173177242279053,
|
| 28 |
+
"eval_runtime": 0.1894,
|
| 29 |
+
"eval_samples_per_second": 31.679,
|
| 30 |
+
"eval_steps_per_second": 10.56,
|
| 31 |
+
"step": 29
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 1.0344827586206897,
|
| 35 |
+
"grad_norm": 16.176136016845703,
|
| 36 |
+
"learning_rate": 1.9408866995073893e-05,
|
| 37 |
+
"loss": 2.7637,
|
| 38 |
+
"step": 30
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.3793103448275863,
|
| 42 |
+
"grad_norm": 5.2406907081604,
|
| 43 |
+
"learning_rate": 1.9211822660098524e-05,
|
| 44 |
+
"loss": 1.0566,
|
| 45 |
+
"step": 40
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.7241379310344827,
|
| 49 |
+
"grad_norm": 3.6556057929992676,
|
| 50 |
+
"learning_rate": 1.9014778325123154e-05,
|
| 51 |
+
"loss": 0.6543,
|
| 52 |
+
"step": 50
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 2.0,
|
| 56 |
+
"eval_loss": 0.26999202370643616,
|
| 57 |
+
"eval_runtime": 0.1963,
|
| 58 |
+
"eval_samples_per_second": 30.571,
|
| 59 |
+
"eval_steps_per_second": 10.19,
|
| 60 |
+
"step": 58
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 2.0689655172413794,
|
| 64 |
+
"grad_norm": 1.8488504886627197,
|
| 65 |
+
"learning_rate": 1.8817733990147784e-05,
|
| 66 |
+
"loss": 0.4713,
|
| 67 |
+
"step": 60
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 2.413793103448276,
|
| 71 |
+
"grad_norm": 2.723362922668457,
|
| 72 |
+
"learning_rate": 1.8620689655172415e-05,
|
| 73 |
+
"loss": 0.3144,
|
| 74 |
+
"step": 70
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.7586206896551726,
|
| 78 |
+
"grad_norm": 1.8972634077072144,
|
| 79 |
+
"learning_rate": 1.8423645320197045e-05,
|
| 80 |
+
"loss": 0.2604,
|
| 81 |
+
"step": 80
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 3.0,
|
| 85 |
+
"eval_loss": 0.09994816780090332,
|
| 86 |
+
"eval_runtime": 0.1912,
|
| 87 |
+
"eval_samples_per_second": 31.381,
|
| 88 |
+
"eval_steps_per_second": 10.46,
|
| 89 |
+
"step": 87
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 3.103448275862069,
|
| 93 |
+
"grad_norm": 3.115511417388916,
|
| 94 |
+
"learning_rate": 1.8226600985221676e-05,
|
| 95 |
+
"loss": 0.2067,
|
| 96 |
+
"step": 90
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 3.4482758620689653,
|
| 100 |
+
"grad_norm": 1.7388259172439575,
|
| 101 |
+
"learning_rate": 1.8029556650246306e-05,
|
| 102 |
+
"loss": 0.1494,
|
| 103 |
+
"step": 100
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 3.793103448275862,
|
| 107 |
+
"grad_norm": 1.2075275182724,
|
| 108 |
+
"learning_rate": 1.7832512315270937e-05,
|
| 109 |
+
"loss": 0.1411,
|
| 110 |
+
"step": 110
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 4.0,
|
| 114 |
+
"eval_loss": 0.05977391079068184,
|
| 115 |
+
"eval_runtime": 0.1933,
|
| 116 |
+
"eval_samples_per_second": 31.034,
|
| 117 |
+
"eval_steps_per_second": 10.345,
|
| 118 |
+
"step": 116
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 4.137931034482759,
|
| 122 |
+
"grad_norm": 1.7329133749008179,
|
| 123 |
+
"learning_rate": 1.7635467980295567e-05,
|
| 124 |
+
"loss": 0.1196,
|
| 125 |
+
"step": 120
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 4.482758620689655,
|
| 129 |
+
"grad_norm": 2.210278272628784,
|
| 130 |
+
"learning_rate": 1.7438423645320198e-05,
|
| 131 |
+
"loss": 0.1132,
|
| 132 |
+
"step": 130
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 4.827586206896552,
|
| 136 |
+
"grad_norm": 1.0056334733963013,
|
| 137 |
+
"learning_rate": 1.7241379310344828e-05,
|
| 138 |
+
"loss": 0.0789,
|
| 139 |
+
"step": 140
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 5.0,
|
| 143 |
+
"eval_loss": 0.04751617833971977,
|
| 144 |
+
"eval_runtime": 0.1898,
|
| 145 |
+
"eval_samples_per_second": 31.605,
|
| 146 |
+
"eval_steps_per_second": 10.535,
|
| 147 |
+
"step": 145
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 5.172413793103448,
|
| 151 |
+
"grad_norm": 0.9742516279220581,
|
| 152 |
+
"learning_rate": 1.704433497536946e-05,
|
| 153 |
+
"loss": 0.0927,
|
| 154 |
+
"step": 150
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 5.517241379310345,
|
| 158 |
+
"grad_norm": 1.4696099758148193,
|
| 159 |
+
"learning_rate": 1.684729064039409e-05,
|
| 160 |
+
"loss": 0.0837,
|
| 161 |
+
"step": 160
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 5.862068965517241,
|
| 165 |
+
"grad_norm": 1.0493124723434448,
|
| 166 |
+
"learning_rate": 1.665024630541872e-05,
|
| 167 |
+
"loss": 0.0689,
|
| 168 |
+
"step": 170
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 6.0,
|
| 172 |
+
"eval_loss": 0.03317258134484291,
|
| 173 |
+
"eval_runtime": 0.2045,
|
| 174 |
+
"eval_samples_per_second": 29.343,
|
| 175 |
+
"eval_steps_per_second": 9.781,
|
| 176 |
+
"step": 174
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 6.206896551724138,
|
| 180 |
+
"grad_norm": 0.9956067204475403,
|
| 181 |
+
"learning_rate": 1.645320197044335e-05,
|
| 182 |
+
"loss": 0.0702,
|
| 183 |
+
"step": 180
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 6.551724137931035,
|
| 187 |
+
"grad_norm": 0.4664933979511261,
|
| 188 |
+
"learning_rate": 1.625615763546798e-05,
|
| 189 |
+
"loss": 0.0675,
|
| 190 |
+
"step": 190
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 6.896551724137931,
|
| 194 |
+
"grad_norm": 1.2444266080856323,
|
| 195 |
+
"learning_rate": 1.605911330049261e-05,
|
| 196 |
+
"loss": 0.0596,
|
| 197 |
+
"step": 200
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 7.0,
|
| 201 |
+
"eval_loss": 0.030222313478589058,
|
| 202 |
+
"eval_runtime": 0.1913,
|
| 203 |
+
"eval_samples_per_second": 31.365,
|
| 204 |
+
"eval_steps_per_second": 10.455,
|
| 205 |
+
"step": 203
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 7.241379310344827,
|
| 209 |
+
"grad_norm": 0.5432140827178955,
|
| 210 |
+
"learning_rate": 1.586206896551724e-05,
|
| 211 |
+
"loss": 0.045,
|
| 212 |
+
"step": 210
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 7.586206896551724,
|
| 216 |
+
"grad_norm": 0.7679450511932373,
|
| 217 |
+
"learning_rate": 1.5665024630541875e-05,
|
| 218 |
+
"loss": 0.0538,
|
| 219 |
+
"step": 220
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 7.931034482758621,
|
| 223 |
+
"grad_norm": 0.7759860754013062,
|
| 224 |
+
"learning_rate": 1.5467980295566506e-05,
|
| 225 |
+
"loss": 0.0624,
|
| 226 |
+
"step": 230
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 8.0,
|
| 230 |
+
"eval_loss": 0.02808324061334133,
|
| 231 |
+
"eval_runtime": 0.2003,
|
| 232 |
+
"eval_samples_per_second": 29.953,
|
| 233 |
+
"eval_steps_per_second": 9.984,
|
| 234 |
+
"step": 232
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 8.275862068965518,
|
| 238 |
+
"grad_norm": 1.7437331676483154,
|
| 239 |
+
"learning_rate": 1.5270935960591133e-05,
|
| 240 |
+
"loss": 0.0369,
|
| 241 |
+
"step": 240
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 8.620689655172415,
|
| 245 |
+
"grad_norm": 0.5273000597953796,
|
| 246 |
+
"learning_rate": 1.5073891625615764e-05,
|
| 247 |
+
"loss": 0.0499,
|
| 248 |
+
"step": 250
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 8.96551724137931,
|
| 252 |
+
"grad_norm": 0.6120426058769226,
|
| 253 |
+
"learning_rate": 1.4876847290640396e-05,
|
| 254 |
+
"loss": 0.0425,
|
| 255 |
+
"step": 260
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 9.0,
|
| 259 |
+
"eval_loss": 0.03035571426153183,
|
| 260 |
+
"eval_runtime": 0.1888,
|
| 261 |
+
"eval_samples_per_second": 31.778,
|
| 262 |
+
"eval_steps_per_second": 10.593,
|
| 263 |
+
"step": 261
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 9.310344827586206,
|
| 267 |
+
"grad_norm": 1.587663173675537,
|
| 268 |
+
"learning_rate": 1.4679802955665026e-05,
|
| 269 |
+
"loss": 0.0395,
|
| 270 |
+
"step": 270
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 9.655172413793103,
|
| 274 |
+
"grad_norm": 0.7260332703590393,
|
| 275 |
+
"learning_rate": 1.4482758620689657e-05,
|
| 276 |
+
"loss": 0.0493,
|
| 277 |
+
"step": 280
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 10.0,
|
| 281 |
+
"grad_norm": 0.9717508554458618,
|
| 282 |
+
"learning_rate": 1.4285714285714287e-05,
|
| 283 |
+
"loss": 0.0424,
|
| 284 |
+
"step": 290
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 10.0,
|
| 288 |
+
"eval_loss": 0.026125147938728333,
|
| 289 |
+
"eval_runtime": 0.0989,
|
| 290 |
+
"eval_samples_per_second": 60.684,
|
| 291 |
+
"eval_steps_per_second": 20.228,
|
| 292 |
+
"step": 290
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 10.344827586206897,
|
| 296 |
+
"grad_norm": 0.7487574815750122,
|
| 297 |
+
"learning_rate": 1.4088669950738918e-05,
|
| 298 |
+
"loss": 0.0412,
|
| 299 |
+
"step": 300
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 10.689655172413794,
|
| 303 |
+
"grad_norm": 1.3717066049575806,
|
| 304 |
+
"learning_rate": 1.3891625615763548e-05,
|
| 305 |
+
"loss": 0.0296,
|
| 306 |
+
"step": 310
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 11.0,
|
| 310 |
+
"eval_loss": 0.031011082231998444,
|
| 311 |
+
"eval_runtime": 0.1922,
|
| 312 |
+
"eval_samples_per_second": 31.22,
|
| 313 |
+
"eval_steps_per_second": 10.407,
|
| 314 |
+
"step": 319
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 11.03448275862069,
|
| 318 |
+
"grad_norm": 0.6860368847846985,
|
| 319 |
+
"learning_rate": 1.369458128078818e-05,
|
| 320 |
+
"loss": 0.0411,
|
| 321 |
+
"step": 320
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 11.379310344827585,
|
| 325 |
+
"grad_norm": 0.9999271035194397,
|
| 326 |
+
"learning_rate": 1.3497536945812807e-05,
|
| 327 |
+
"loss": 0.0358,
|
| 328 |
+
"step": 330
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 11.724137931034482,
|
| 332 |
+
"grad_norm": 1.2313721179962158,
|
| 333 |
+
"learning_rate": 1.330049261083744e-05,
|
| 334 |
+
"loss": 0.0324,
|
| 335 |
+
"step": 340
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 12.0,
|
| 339 |
+
"eval_loss": 0.029439905658364296,
|
| 340 |
+
"eval_runtime": 0.2015,
|
| 341 |
+
"eval_samples_per_second": 29.784,
|
| 342 |
+
"eval_steps_per_second": 9.928,
|
| 343 |
+
"step": 348
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 12.068965517241379,
|
| 347 |
+
"grad_norm": 0.9032502174377441,
|
| 348 |
+
"learning_rate": 1.310344827586207e-05,
|
| 349 |
+
"loss": 0.041,
|
| 350 |
+
"step": 350
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 12.413793103448276,
|
| 354 |
+
"grad_norm": 0.612800657749176,
|
| 355 |
+
"learning_rate": 1.29064039408867e-05,
|
| 356 |
+
"loss": 0.0384,
|
| 357 |
+
"step": 360
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 12.758620689655173,
|
| 361 |
+
"grad_norm": 0.19512540102005005,
|
| 362 |
+
"learning_rate": 1.2709359605911331e-05,
|
| 363 |
+
"loss": 0.0349,
|
| 364 |
+
"step": 370
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 13.0,
|
| 368 |
+
"eval_loss": 0.02956104278564453,
|
| 369 |
+
"eval_runtime": 0.1909,
|
| 370 |
+
"eval_samples_per_second": 31.424,
|
| 371 |
+
"eval_steps_per_second": 10.475,
|
| 372 |
+
"step": 377
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 13.10344827586207,
|
| 376 |
+
"grad_norm": 0.8481155633926392,
|
| 377 |
+
"learning_rate": 1.2512315270935961e-05,
|
| 378 |
+
"loss": 0.0271,
|
| 379 |
+
"step": 380
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 13.448275862068966,
|
| 383 |
+
"grad_norm": 1.3683249950408936,
|
| 384 |
+
"learning_rate": 1.2315270935960592e-05,
|
| 385 |
+
"loss": 0.0265,
|
| 386 |
+
"step": 390
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 13.793103448275861,
|
| 390 |
+
"grad_norm": 0.6365839838981628,
|
| 391 |
+
"learning_rate": 1.2118226600985224e-05,
|
| 392 |
+
"loss": 0.0298,
|
| 393 |
+
"step": 400
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 14.0,
|
| 397 |
+
"eval_loss": 0.03038203716278076,
|
| 398 |
+
"eval_runtime": 0.1902,
|
| 399 |
+
"eval_samples_per_second": 31.538,
|
| 400 |
+
"eval_steps_per_second": 10.513,
|
| 401 |
+
"step": 406
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 14.137931034482758,
|
| 405 |
+
"grad_norm": 0.7672634124755859,
|
| 406 |
+
"learning_rate": 1.1921182266009855e-05,
|
| 407 |
+
"loss": 0.0335,
|
| 408 |
+
"step": 410
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 14.482758620689655,
|
| 412 |
+
"grad_norm": 0.2541676163673401,
|
| 413 |
+
"learning_rate": 1.1724137931034483e-05,
|
| 414 |
+
"loss": 0.0286,
|
| 415 |
+
"step": 420
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 14.827586206896552,
|
| 419 |
+
"grad_norm": 0.8434980511665344,
|
| 420 |
+
"learning_rate": 1.1527093596059114e-05,
|
| 421 |
+
"loss": 0.0205,
|
| 422 |
+
"step": 430
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 15.0,
|
| 426 |
+
"eval_loss": 0.030394822359085083,
|
| 427 |
+
"eval_runtime": 0.19,
|
| 428 |
+
"eval_samples_per_second": 31.587,
|
| 429 |
+
"eval_steps_per_second": 10.529,
|
| 430 |
+
"step": 435
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 15.172413793103448,
|
| 434 |
+
"grad_norm": 0.4303562641143799,
|
| 435 |
+
"learning_rate": 1.1330049261083744e-05,
|
| 436 |
+
"loss": 0.0323,
|
| 437 |
+
"step": 440
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 15.517241379310345,
|
| 441 |
+
"grad_norm": 0.42710408568382263,
|
| 442 |
+
"learning_rate": 1.1133004926108375e-05,
|
| 443 |
+
"loss": 0.0227,
|
| 444 |
+
"step": 450
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 15.862068965517242,
|
| 448 |
+
"grad_norm": 0.6126664876937866,
|
| 449 |
+
"learning_rate": 1.0935960591133005e-05,
|
| 450 |
+
"loss": 0.0215,
|
| 451 |
+
"step": 460
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 16.0,
|
| 455 |
+
"eval_loss": 0.030341001227498055,
|
| 456 |
+
"eval_runtime": 0.1895,
|
| 457 |
+
"eval_samples_per_second": 31.669,
|
| 458 |
+
"eval_steps_per_second": 10.556,
|
| 459 |
+
"step": 464
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 16.20689655172414,
|
| 463 |
+
"grad_norm": 0.7011211514472961,
|
| 464 |
+
"learning_rate": 1.0738916256157637e-05,
|
| 465 |
+
"loss": 0.0337,
|
| 466 |
+
"step": 470
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 16.551724137931036,
|
| 470 |
+
"grad_norm": 0.5126092433929443,
|
| 471 |
+
"learning_rate": 1.0541871921182268e-05,
|
| 472 |
+
"loss": 0.0233,
|
| 473 |
+
"step": 480
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 16.896551724137932,
|
| 477 |
+
"grad_norm": 0.9033933281898499,
|
| 478 |
+
"learning_rate": 1.0344827586206898e-05,
|
| 479 |
+
"loss": 0.0182,
|
| 480 |
+
"step": 490
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 17.0,
|
| 484 |
+
"eval_loss": 0.027958964928984642,
|
| 485 |
+
"eval_runtime": 0.1902,
|
| 486 |
+
"eval_samples_per_second": 31.553,
|
| 487 |
+
"eval_steps_per_second": 10.518,
|
| 488 |
+
"step": 493
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 17.24137931034483,
|
| 492 |
+
"grad_norm": 0.4326847493648529,
|
| 493 |
+
"learning_rate": 1.0147783251231529e-05,
|
| 494 |
+
"loss": 0.0249,
|
| 495 |
+
"step": 500
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 17.586206896551722,
|
| 499 |
+
"grad_norm": 0.4034535586833954,
|
| 500 |
+
"learning_rate": 9.95073891625616e-06,
|
| 501 |
+
"loss": 0.0215,
|
| 502 |
+
"step": 510
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 17.93103448275862,
|
| 506 |
+
"grad_norm": 0.4999659061431885,
|
| 507 |
+
"learning_rate": 9.75369458128079e-06,
|
| 508 |
+
"loss": 0.0247,
|
| 509 |
+
"step": 520
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 18.0,
|
| 513 |
+
"eval_loss": 0.030091799795627594,
|
| 514 |
+
"eval_runtime": 0.1897,
|
| 515 |
+
"eval_samples_per_second": 31.624,
|
| 516 |
+
"eval_steps_per_second": 10.541,
|
| 517 |
+
"step": 522
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 18.275862068965516,
|
| 521 |
+
"grad_norm": 0.3004280924797058,
|
| 522 |
+
"learning_rate": 9.55665024630542e-06,
|
| 523 |
+
"loss": 0.0161,
|
| 524 |
+
"step": 530
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 18.620689655172413,
|
| 528 |
+
"grad_norm": 0.5018593072891235,
|
| 529 |
+
"learning_rate": 9.359605911330049e-06,
|
| 530 |
+
"loss": 0.0227,
|
| 531 |
+
"step": 540
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 18.96551724137931,
|
| 535 |
+
"grad_norm": 0.5728505849838257,
|
| 536 |
+
"learning_rate": 9.162561576354681e-06,
|
| 537 |
+
"loss": 0.0221,
|
| 538 |
+
"step": 550
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 19.0,
|
| 542 |
+
"eval_loss": 0.030765995383262634,
|
| 543 |
+
"eval_runtime": 0.1882,
|
| 544 |
+
"eval_samples_per_second": 31.886,
|
| 545 |
+
"eval_steps_per_second": 10.629,
|
| 546 |
+
"step": 551
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 19.310344827586206,
|
| 550 |
+
"grad_norm": 1.1493581533432007,
|
| 551 |
+
"learning_rate": 8.965517241379312e-06,
|
| 552 |
+
"loss": 0.0204,
|
| 553 |
+
"step": 560
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 19.655172413793103,
|
| 557 |
+
"grad_norm": 0.3401670753955841,
|
| 558 |
+
"learning_rate": 8.768472906403942e-06,
|
| 559 |
+
"loss": 0.0164,
|
| 560 |
+
"step": 570
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 20.0,
|
| 564 |
+
"grad_norm": 0.9875850081443787,
|
| 565 |
+
"learning_rate": 8.571428571428571e-06,
|
| 566 |
+
"loss": 0.0213,
|
| 567 |
+
"step": 580
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 20.0,
|
| 571 |
+
"eval_loss": 0.028663409873843193,
|
| 572 |
+
"eval_runtime": 0.1008,
|
| 573 |
+
"eval_samples_per_second": 59.511,
|
| 574 |
+
"eval_steps_per_second": 19.837,
|
| 575 |
+
"step": 580
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 20.344827586206897,
|
| 579 |
+
"grad_norm": 0.6725947260856628,
|
| 580 |
+
"learning_rate": 8.374384236453203e-06,
|
| 581 |
+
"loss": 0.0192,
|
| 582 |
+
"step": 590
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 20.689655172413794,
|
| 586 |
+
"grad_norm": 0.6594141721725464,
|
| 587 |
+
"learning_rate": 8.177339901477834e-06,
|
| 588 |
+
"loss": 0.0194,
|
| 589 |
+
"step": 600
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 21.0,
|
| 593 |
+
"eval_loss": 0.027452999725937843,
|
| 594 |
+
"eval_runtime": 0.1899,
|
| 595 |
+
"eval_samples_per_second": 31.602,
|
| 596 |
+
"eval_steps_per_second": 10.534,
|
| 597 |
+
"step": 609
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 21.03448275862069,
|
| 601 |
+
"grad_norm": 0.32469141483306885,
|
| 602 |
+
"learning_rate": 7.980295566502464e-06,
|
| 603 |
+
"loss": 0.0167,
|
| 604 |
+
"step": 610
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 21.379310344827587,
|
| 608 |
+
"grad_norm": 0.7029064893722534,
|
| 609 |
+
"learning_rate": 7.783251231527095e-06,
|
| 610 |
+
"loss": 0.0237,
|
| 611 |
+
"step": 620
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 21.724137931034484,
|
| 615 |
+
"grad_norm": 0.5634991526603699,
|
| 616 |
+
"learning_rate": 7.586206896551724e-06,
|
| 617 |
+
"loss": 0.0212,
|
| 618 |
+
"step": 630
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 22.0,
|
| 622 |
+
"eval_loss": 0.027833983302116394,
|
| 623 |
+
"eval_runtime": 0.1913,
|
| 624 |
+
"eval_samples_per_second": 31.357,
|
| 625 |
+
"eval_steps_per_second": 10.452,
|
| 626 |
+
"step": 638
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 22.06896551724138,
|
| 630 |
+
"grad_norm": 0.7171387672424316,
|
| 631 |
+
"learning_rate": 7.3891625615763555e-06,
|
| 632 |
+
"loss": 0.0204,
|
| 633 |
+
"step": 640
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 22.413793103448278,
|
| 637 |
+
"grad_norm": 0.3534681797027588,
|
| 638 |
+
"learning_rate": 7.192118226600986e-06,
|
| 639 |
+
"loss": 0.0162,
|
| 640 |
+
"step": 650
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 22.75862068965517,
|
| 644 |
+
"grad_norm": 0.2446085512638092,
|
| 645 |
+
"learning_rate": 6.995073891625616e-06,
|
| 646 |
+
"loss": 0.0178,
|
| 647 |
+
"step": 660
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 23.0,
|
| 651 |
+
"eval_loss": 0.0313410721719265,
|
| 652 |
+
"eval_runtime": 0.1923,
|
| 653 |
+
"eval_samples_per_second": 31.2,
|
| 654 |
+
"eval_steps_per_second": 10.4,
|
| 655 |
+
"step": 667
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 23.103448275862068,
|
| 659 |
+
"grad_norm": 0.3782992959022522,
|
| 660 |
+
"learning_rate": 6.798029556650246e-06,
|
| 661 |
+
"loss": 0.0214,
|
| 662 |
+
"step": 670
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 23.448275862068964,
|
| 666 |
+
"grad_norm": 0.2016129344701767,
|
| 667 |
+
"learning_rate": 6.600985221674877e-06,
|
| 668 |
+
"loss": 0.0167,
|
| 669 |
+
"step": 680
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 23.79310344827586,
|
| 673 |
+
"grad_norm": 0.3993048667907715,
|
| 674 |
+
"learning_rate": 6.403940886699508e-06,
|
| 675 |
+
"loss": 0.0217,
|
| 676 |
+
"step": 690
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 24.0,
|
| 680 |
+
"eval_loss": 0.030709436163306236,
|
| 681 |
+
"eval_runtime": 0.1914,
|
| 682 |
+
"eval_samples_per_second": 31.345,
|
| 683 |
+
"eval_steps_per_second": 10.448,
|
| 684 |
+
"step": 696
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 24.137931034482758,
|
| 688 |
+
"grad_norm": 0.7120524048805237,
|
| 689 |
+
"learning_rate": 6.206896551724138e-06,
|
| 690 |
+
"loss": 0.0227,
|
| 691 |
+
"step": 700
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 24.482758620689655,
|
| 695 |
+
"grad_norm": 0.3477053940296173,
|
| 696 |
+
"learning_rate": 6.00985221674877e-06,
|
| 697 |
+
"loss": 0.0203,
|
| 698 |
+
"step": 710
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 24.82758620689655,
|
| 702 |
+
"grad_norm": 0.30051225423812866,
|
| 703 |
+
"learning_rate": 5.812807881773399e-06,
|
| 704 |
+
"loss": 0.0118,
|
| 705 |
+
"step": 720
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 25.0,
|
| 709 |
+
"eval_loss": 0.02974347025156021,
|
| 710 |
+
"eval_runtime": 0.1936,
|
| 711 |
+
"eval_samples_per_second": 30.992,
|
| 712 |
+
"eval_steps_per_second": 10.331,
|
| 713 |
+
"step": 725
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 25.17241379310345,
|
| 717 |
+
"grad_norm": 0.38762152194976807,
|
| 718 |
+
"learning_rate": 5.61576354679803e-06,
|
| 719 |
+
"loss": 0.0161,
|
| 720 |
+
"step": 730
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 25.517241379310345,
|
| 724 |
+
"grad_norm": 0.48786938190460205,
|
| 725 |
+
"learning_rate": 5.41871921182266e-06,
|
| 726 |
+
"loss": 0.0153,
|
| 727 |
+
"step": 740
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 25.862068965517242,
|
| 731 |
+
"grad_norm": 0.273346483707428,
|
| 732 |
+
"learning_rate": 5.2216748768472915e-06,
|
| 733 |
+
"loss": 0.0189,
|
| 734 |
+
"step": 750
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 26.0,
|
| 738 |
+
"eval_loss": 0.031065121293067932,
|
| 739 |
+
"eval_runtime": 0.1912,
|
| 740 |
+
"eval_samples_per_second": 31.387,
|
| 741 |
+
"eval_steps_per_second": 10.462,
|
| 742 |
+
"step": 754
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 26.20689655172414,
|
| 746 |
+
"grad_norm": 0.8301162123680115,
|
| 747 |
+
"learning_rate": 5.024630541871922e-06,
|
| 748 |
+
"loss": 0.0194,
|
| 749 |
+
"step": 760
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 26.551724137931036,
|
| 753 |
+
"grad_norm": 0.46895724534988403,
|
| 754 |
+
"learning_rate": 4.8275862068965525e-06,
|
| 755 |
+
"loss": 0.0149,
|
| 756 |
+
"step": 770
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 26.896551724137932,
|
| 760 |
+
"grad_norm": 0.4784580171108246,
|
| 761 |
+
"learning_rate": 4.630541871921182e-06,
|
| 762 |
+
"loss": 0.0185,
|
| 763 |
+
"step": 780
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 27.0,
|
| 767 |
+
"eval_loss": 0.029767701402306557,
|
| 768 |
+
"eval_runtime": 0.1921,
|
| 769 |
+
"eval_samples_per_second": 31.24,
|
| 770 |
+
"eval_steps_per_second": 10.413,
|
| 771 |
+
"step": 783
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 27.24137931034483,
|
| 775 |
+
"grad_norm": 0.7427147626876831,
|
| 776 |
+
"learning_rate": 4.4334975369458135e-06,
|
| 777 |
+
"loss": 0.0155,
|
| 778 |
+
"step": 790
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 27.586206896551722,
|
| 782 |
+
"grad_norm": 0.4144653081893921,
|
| 783 |
+
"learning_rate": 4.236453201970444e-06,
|
| 784 |
+
"loss": 0.0159,
|
| 785 |
+
"step": 800
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 27.93103448275862,
|
| 789 |
+
"grad_norm": 0.29942747950553894,
|
| 790 |
+
"learning_rate": 4.039408866995074e-06,
|
| 791 |
+
"loss": 0.021,
|
| 792 |
+
"step": 810
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 28.0,
|
| 796 |
+
"eval_loss": 0.029253564774990082,
|
| 797 |
+
"eval_runtime": 0.1914,
|
| 798 |
+
"eval_samples_per_second": 31.352,
|
| 799 |
+
"eval_steps_per_second": 10.451,
|
| 800 |
+
"step": 812
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 28.275862068965516,
|
| 804 |
+
"grad_norm": 0.21119730174541473,
|
| 805 |
+
"learning_rate": 3.842364532019705e-06,
|
| 806 |
+
"loss": 0.0153,
|
| 807 |
+
"step": 820
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 28.620689655172413,
|
| 811 |
+
"grad_norm": 0.2980373501777649,
|
| 812 |
+
"learning_rate": 3.6453201970443354e-06,
|
| 813 |
+
"loss": 0.0189,
|
| 814 |
+
"step": 830
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 28.96551724137931,
|
| 818 |
+
"grad_norm": 0.09113238751888275,
|
| 819 |
+
"learning_rate": 3.448275862068966e-06,
|
| 820 |
+
"loss": 0.0102,
|
| 821 |
+
"step": 840
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 29.0,
|
| 825 |
+
"eval_loss": 0.028439467772841454,
|
| 826 |
+
"eval_runtime": 0.1908,
|
| 827 |
+
"eval_samples_per_second": 31.449,
|
| 828 |
+
"eval_steps_per_second": 10.483,
|
| 829 |
+
"step": 841
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 29.310344827586206,
|
| 833 |
+
"grad_norm": 0.3905262351036072,
|
| 834 |
+
"learning_rate": 3.2512315270935963e-06,
|
| 835 |
+
"loss": 0.0153,
|
| 836 |
+
"step": 850
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 29.655172413793103,
|
| 840 |
+
"grad_norm": 0.36752381920814514,
|
| 841 |
+
"learning_rate": 3.054187192118227e-06,
|
| 842 |
+
"loss": 0.014,
|
| 843 |
+
"step": 860
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 30.0,
|
| 847 |
+
"grad_norm": 0.6769506931304932,
|
| 848 |
+
"learning_rate": 2.8571428571428573e-06,
|
| 849 |
+
"loss": 0.018,
|
| 850 |
+
"step": 870
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 30.0,
|
| 854 |
+
"eval_loss": 0.02924039028584957,
|
| 855 |
+
"eval_runtime": 0.102,
|
| 856 |
+
"eval_samples_per_second": 58.797,
|
| 857 |
+
"eval_steps_per_second": 19.599,
|
| 858 |
+
"step": 870
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 30.344827586206897,
|
| 862 |
+
"grad_norm": 0.676186740398407,
|
| 863 |
+
"learning_rate": 2.660098522167488e-06,
|
| 864 |
+
"loss": 0.0202,
|
| 865 |
+
"step": 880
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 30.689655172413794,
|
| 869 |
+
"grad_norm": 0.2909271717071533,
|
| 870 |
+
"learning_rate": 2.4630541871921186e-06,
|
| 871 |
+
"loss": 0.0105,
|
| 872 |
+
"step": 890
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 31.0,
|
| 876 |
+
"eval_loss": 0.030303308740258217,
|
| 877 |
+
"eval_runtime": 0.191,
|
| 878 |
+
"eval_samples_per_second": 31.415,
|
| 879 |
+
"eval_steps_per_second": 10.472,
|
| 880 |
+
"step": 899
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 31.03448275862069,
|
| 884 |
+
"grad_norm": 0.8155515193939209,
|
| 885 |
+
"learning_rate": 2.266009852216749e-06,
|
| 886 |
+
"loss": 0.0174,
|
| 887 |
+
"step": 900
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 31.379310344827587,
|
| 891 |
+
"grad_norm": 0.5190662741661072,
|
| 892 |
+
"learning_rate": 2.0689655172413796e-06,
|
| 893 |
+
"loss": 0.0157,
|
| 894 |
+
"step": 910
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 31.724137931034484,
|
| 898 |
+
"grad_norm": 0.8997210264205933,
|
| 899 |
+
"learning_rate": 1.8719211822660098e-06,
|
| 900 |
+
"loss": 0.0155,
|
| 901 |
+
"step": 920
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 32.0,
|
| 905 |
+
"eval_loss": 0.030399195849895477,
|
| 906 |
+
"eval_runtime": 0.1899,
|
| 907 |
+
"eval_samples_per_second": 31.593,
|
| 908 |
+
"eval_steps_per_second": 10.531,
|
| 909 |
+
"step": 928
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 32.06896551724138,
|
| 913 |
+
"grad_norm": 0.5701755881309509,
|
| 914 |
+
"learning_rate": 1.6748768472906405e-06,
|
| 915 |
+
"loss": 0.0139,
|
| 916 |
+
"step": 930
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 32.41379310344828,
|
| 920 |
+
"grad_norm": 0.16824859380722046,
|
| 921 |
+
"learning_rate": 1.4778325123152712e-06,
|
| 922 |
+
"loss": 0.0113,
|
| 923 |
+
"step": 940
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 32.758620689655174,
|
| 927 |
+
"grad_norm": 0.5066978931427002,
|
| 928 |
+
"learning_rate": 1.2807881773399017e-06,
|
| 929 |
+
"loss": 0.0127,
|
| 930 |
+
"step": 950
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 33.0,
|
| 934 |
+
"eval_loss": 0.030721982941031456,
|
| 935 |
+
"eval_runtime": 0.191,
|
| 936 |
+
"eval_samples_per_second": 31.41,
|
| 937 |
+
"eval_steps_per_second": 10.47,
|
| 938 |
+
"step": 957
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 33.10344827586207,
|
| 942 |
+
"grad_norm": 0.5038828253746033,
|
| 943 |
+
"learning_rate": 1.0837438423645322e-06,
|
| 944 |
+
"loss": 0.0138,
|
| 945 |
+
"step": 960
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 33.44827586206897,
|
| 949 |
+
"grad_norm": 0.4406326115131378,
|
| 950 |
+
"learning_rate": 8.866995073891626e-07,
|
| 951 |
+
"loss": 0.0116,
|
| 952 |
+
"step": 970
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 33.793103448275865,
|
| 956 |
+
"grad_norm": 0.3732337951660156,
|
| 957 |
+
"learning_rate": 6.896551724137931e-07,
|
| 958 |
+
"loss": 0.0171,
|
| 959 |
+
"step": 980
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 34.0,
|
| 963 |
+
"eval_loss": 0.03042110800743103,
|
| 964 |
+
"eval_runtime": 0.191,
|
| 965 |
+
"eval_samples_per_second": 31.415,
|
| 966 |
+
"eval_steps_per_second": 10.472,
|
| 967 |
+
"step": 986
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 34.13793103448276,
|
| 971 |
+
"grad_norm": 0.32739463448524475,
|
| 972 |
+
"learning_rate": 4.926108374384237e-07,
|
| 973 |
+
"loss": 0.0186,
|
| 974 |
+
"step": 990
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 34.48275862068966,
|
| 978 |
+
"grad_norm": 0.45314541459083557,
|
| 979 |
+
"learning_rate": 2.955665024630542e-07,
|
| 980 |
+
"loss": 0.0108,
|
| 981 |
+
"step": 1000
|
| 982 |
+
}
|
| 983 |
+
],
|
| 984 |
+
"logging_steps": 10,
|
| 985 |
+
"max_steps": 1015,
|
| 986 |
+
"num_input_tokens_seen": 0,
|
| 987 |
+
"num_train_epochs": 35,
|
| 988 |
+
"save_steps": 500,
|
| 989 |
+
"stateful_callbacks": {
|
| 990 |
+
"TrainerControl": {
|
| 991 |
+
"args": {
|
| 992 |
+
"should_epoch_stop": false,
|
| 993 |
+
"should_evaluate": false,
|
| 994 |
+
"should_log": false,
|
| 995 |
+
"should_save": true,
|
| 996 |
+
"should_training_stop": false
|
| 997 |
+
},
|
| 998 |
+
"attributes": {}
|
| 999 |
+
}
|
| 1000 |
+
},
|
| 1001 |
+
"total_flos": 2128241491968000.0,
|
| 1002 |
+
"train_batch_size": 4,
|
| 1003 |
+
"trial_name": null,
|
| 1004 |
+
"trial_params": null
|
| 1005 |
+
}
|
json_extraction_point_activity/checkpoint-1000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:333cc851b675d7a01620eeba59cdfbc1a624d74927d086ab660bd04fef9b6029
|
| 3 |
+
size 5240
|
json_extraction_point_activity/checkpoint-1015/config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "t5-large",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"T5ForConditionalGeneration"
|
| 5 |
+
],
|
| 6 |
+
"classifier_dropout": 0.0,
|
| 7 |
+
"d_ff": 4096,
|
| 8 |
+
"d_kv": 64,
|
| 9 |
+
"d_model": 1024,
|
| 10 |
+
"decoder_start_token_id": 0,
|
| 11 |
+
"dense_act_fn": "relu",
|
| 12 |
+
"dropout_rate": 0.1,
|
| 13 |
+
"eos_token_id": 1,
|
| 14 |
+
"feed_forward_proj": "relu",
|
| 15 |
+
"initializer_factor": 1.0,
|
| 16 |
+
"is_encoder_decoder": true,
|
| 17 |
+
"is_gated_act": false,
|
| 18 |
+
"layer_norm_epsilon": 1e-06,
|
| 19 |
+
"model_type": "t5",
|
| 20 |
+
"n_positions": 512,
|
| 21 |
+
"num_decoder_layers": 24,
|
| 22 |
+
"num_heads": 16,
|
| 23 |
+
"num_layers": 24,
|
| 24 |
+
"output_past": true,
|
| 25 |
+
"pad_token_id": 0,
|
| 26 |
+
"relative_attention_max_distance": 128,
|
| 27 |
+
"relative_attention_num_buckets": 32,
|
| 28 |
+
"task_specific_params": {
|
| 29 |
+
"summarization": {
|
| 30 |
+
"early_stopping": true,
|
| 31 |
+
"length_penalty": 2.0,
|
| 32 |
+
"max_length": 200,
|
| 33 |
+
"min_length": 30,
|
| 34 |
+
"no_repeat_ngram_size": 3,
|
| 35 |
+
"num_beams": 4,
|
| 36 |
+
"prefix": "summarize: "
|
| 37 |
+
},
|
| 38 |
+
"translation_en_to_de": {
|
| 39 |
+
"early_stopping": true,
|
| 40 |
+
"max_length": 300,
|
| 41 |
+
"num_beams": 4,
|
| 42 |
+
"prefix": "translate English to German: "
|
| 43 |
+
},
|
| 44 |
+
"translation_en_to_fr": {
|
| 45 |
+
"early_stopping": true,
|
| 46 |
+
"max_length": 300,
|
| 47 |
+
"num_beams": 4,
|
| 48 |
+
"prefix": "translate English to French: "
|
| 49 |
+
},
|
| 50 |
+
"translation_en_to_ro": {
|
| 51 |
+
"early_stopping": true,
|
| 52 |
+
"max_length": 300,
|
| 53 |
+
"num_beams": 4,
|
| 54 |
+
"prefix": "translate English to Romanian: "
|
| 55 |
+
}
|
| 56 |
+
},
|
| 57 |
+
"torch_dtype": "float32",
|
| 58 |
+
"transformers_version": "4.48.2",
|
| 59 |
+
"use_cache": true,
|
| 60 |
+
"vocab_size": 32128
|
| 61 |
+
}
|
json_extraction_point_activity/checkpoint-1015/generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"decoder_start_token_id": 0,
|
| 4 |
+
"eos_token_id": 1,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.48.2"
|
| 7 |
+
}
|
json_extraction_point_activity/checkpoint-1015/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a3f3216f61837c04638fd20747110cf419cd0cf7777e19c57b1a79ccbd5cce8
|
| 3 |
+
size 2950734544
|
json_extraction_point_activity/checkpoint-1015/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5f50e5090f6f75af54d40ad57109be3f0f7daaffacb86271dba6ac32dfc076bb
|
| 3 |
+
size 5901778825
|
json_extraction_point_activity/checkpoint-1015/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a2a4fa7b4c0bb14dd8ab49891754dfc89b075c92267608720041d0fd455797e7
|
| 3 |
+
size 14244
|
json_extraction_point_activity/checkpoint-1015/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f2ccfc7b23418fc7ce59db86126d54164c357ffeea106e4ff6ff9ed8cf664181
|
| 3 |
+
size 1064
|
json_extraction_point_activity/checkpoint-1015/trainer_state.json
ADDED
|
@@ -0,0 +1,1012 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 35.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1015,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.3448275862068966,
|
| 13 |
+
"grad_norm": 174.2040557861328,
|
| 14 |
+
"learning_rate": 1.9802955665024632e-05,
|
| 15 |
+
"loss": 11.9232,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.6896551724137931,
|
| 20 |
+
"grad_norm": 54.92245101928711,
|
| 21 |
+
"learning_rate": 1.9605911330049263e-05,
|
| 22 |
+
"loss": 6.0635,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 1.0,
|
| 27 |
+
"eval_loss": 0.8173177242279053,
|
| 28 |
+
"eval_runtime": 0.1894,
|
| 29 |
+
"eval_samples_per_second": 31.679,
|
| 30 |
+
"eval_steps_per_second": 10.56,
|
| 31 |
+
"step": 29
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 1.0344827586206897,
|
| 35 |
+
"grad_norm": 16.176136016845703,
|
| 36 |
+
"learning_rate": 1.9408866995073893e-05,
|
| 37 |
+
"loss": 2.7637,
|
| 38 |
+
"step": 30
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.3793103448275863,
|
| 42 |
+
"grad_norm": 5.2406907081604,
|
| 43 |
+
"learning_rate": 1.9211822660098524e-05,
|
| 44 |
+
"loss": 1.0566,
|
| 45 |
+
"step": 40
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.7241379310344827,
|
| 49 |
+
"grad_norm": 3.6556057929992676,
|
| 50 |
+
"learning_rate": 1.9014778325123154e-05,
|
| 51 |
+
"loss": 0.6543,
|
| 52 |
+
"step": 50
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 2.0,
|
| 56 |
+
"eval_loss": 0.26999202370643616,
|
| 57 |
+
"eval_runtime": 0.1963,
|
| 58 |
+
"eval_samples_per_second": 30.571,
|
| 59 |
+
"eval_steps_per_second": 10.19,
|
| 60 |
+
"step": 58
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 2.0689655172413794,
|
| 64 |
+
"grad_norm": 1.8488504886627197,
|
| 65 |
+
"learning_rate": 1.8817733990147784e-05,
|
| 66 |
+
"loss": 0.4713,
|
| 67 |
+
"step": 60
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 2.413793103448276,
|
| 71 |
+
"grad_norm": 2.723362922668457,
|
| 72 |
+
"learning_rate": 1.8620689655172415e-05,
|
| 73 |
+
"loss": 0.3144,
|
| 74 |
+
"step": 70
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.7586206896551726,
|
| 78 |
+
"grad_norm": 1.8972634077072144,
|
| 79 |
+
"learning_rate": 1.8423645320197045e-05,
|
| 80 |
+
"loss": 0.2604,
|
| 81 |
+
"step": 80
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 3.0,
|
| 85 |
+
"eval_loss": 0.09994816780090332,
|
| 86 |
+
"eval_runtime": 0.1912,
|
| 87 |
+
"eval_samples_per_second": 31.381,
|
| 88 |
+
"eval_steps_per_second": 10.46,
|
| 89 |
+
"step": 87
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 3.103448275862069,
|
| 93 |
+
"grad_norm": 3.115511417388916,
|
| 94 |
+
"learning_rate": 1.8226600985221676e-05,
|
| 95 |
+
"loss": 0.2067,
|
| 96 |
+
"step": 90
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 3.4482758620689653,
|
| 100 |
+
"grad_norm": 1.7388259172439575,
|
| 101 |
+
"learning_rate": 1.8029556650246306e-05,
|
| 102 |
+
"loss": 0.1494,
|
| 103 |
+
"step": 100
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 3.793103448275862,
|
| 107 |
+
"grad_norm": 1.2075275182724,
|
| 108 |
+
"learning_rate": 1.7832512315270937e-05,
|
| 109 |
+
"loss": 0.1411,
|
| 110 |
+
"step": 110
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 4.0,
|
| 114 |
+
"eval_loss": 0.05977391079068184,
|
| 115 |
+
"eval_runtime": 0.1933,
|
| 116 |
+
"eval_samples_per_second": 31.034,
|
| 117 |
+
"eval_steps_per_second": 10.345,
|
| 118 |
+
"step": 116
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 4.137931034482759,
|
| 122 |
+
"grad_norm": 1.7329133749008179,
|
| 123 |
+
"learning_rate": 1.7635467980295567e-05,
|
| 124 |
+
"loss": 0.1196,
|
| 125 |
+
"step": 120
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 4.482758620689655,
|
| 129 |
+
"grad_norm": 2.210278272628784,
|
| 130 |
+
"learning_rate": 1.7438423645320198e-05,
|
| 131 |
+
"loss": 0.1132,
|
| 132 |
+
"step": 130
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 4.827586206896552,
|
| 136 |
+
"grad_norm": 1.0056334733963013,
|
| 137 |
+
"learning_rate": 1.7241379310344828e-05,
|
| 138 |
+
"loss": 0.0789,
|
| 139 |
+
"step": 140
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 5.0,
|
| 143 |
+
"eval_loss": 0.04751617833971977,
|
| 144 |
+
"eval_runtime": 0.1898,
|
| 145 |
+
"eval_samples_per_second": 31.605,
|
| 146 |
+
"eval_steps_per_second": 10.535,
|
| 147 |
+
"step": 145
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 5.172413793103448,
|
| 151 |
+
"grad_norm": 0.9742516279220581,
|
| 152 |
+
"learning_rate": 1.704433497536946e-05,
|
| 153 |
+
"loss": 0.0927,
|
| 154 |
+
"step": 150
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 5.517241379310345,
|
| 158 |
+
"grad_norm": 1.4696099758148193,
|
| 159 |
+
"learning_rate": 1.684729064039409e-05,
|
| 160 |
+
"loss": 0.0837,
|
| 161 |
+
"step": 160
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 5.862068965517241,
|
| 165 |
+
"grad_norm": 1.0493124723434448,
|
| 166 |
+
"learning_rate": 1.665024630541872e-05,
|
| 167 |
+
"loss": 0.0689,
|
| 168 |
+
"step": 170
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 6.0,
|
| 172 |
+
"eval_loss": 0.03317258134484291,
|
| 173 |
+
"eval_runtime": 0.2045,
|
| 174 |
+
"eval_samples_per_second": 29.343,
|
| 175 |
+
"eval_steps_per_second": 9.781,
|
| 176 |
+
"step": 174
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 6.206896551724138,
|
| 180 |
+
"grad_norm": 0.9956067204475403,
|
| 181 |
+
"learning_rate": 1.645320197044335e-05,
|
| 182 |
+
"loss": 0.0702,
|
| 183 |
+
"step": 180
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 6.551724137931035,
|
| 187 |
+
"grad_norm": 0.4664933979511261,
|
| 188 |
+
"learning_rate": 1.625615763546798e-05,
|
| 189 |
+
"loss": 0.0675,
|
| 190 |
+
"step": 190
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 6.896551724137931,
|
| 194 |
+
"grad_norm": 1.2444266080856323,
|
| 195 |
+
"learning_rate": 1.605911330049261e-05,
|
| 196 |
+
"loss": 0.0596,
|
| 197 |
+
"step": 200
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 7.0,
|
| 201 |
+
"eval_loss": 0.030222313478589058,
|
| 202 |
+
"eval_runtime": 0.1913,
|
| 203 |
+
"eval_samples_per_second": 31.365,
|
| 204 |
+
"eval_steps_per_second": 10.455,
|
| 205 |
+
"step": 203
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 7.241379310344827,
|
| 209 |
+
"grad_norm": 0.5432140827178955,
|
| 210 |
+
"learning_rate": 1.586206896551724e-05,
|
| 211 |
+
"loss": 0.045,
|
| 212 |
+
"step": 210
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 7.586206896551724,
|
| 216 |
+
"grad_norm": 0.7679450511932373,
|
| 217 |
+
"learning_rate": 1.5665024630541875e-05,
|
| 218 |
+
"loss": 0.0538,
|
| 219 |
+
"step": 220
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 7.931034482758621,
|
| 223 |
+
"grad_norm": 0.7759860754013062,
|
| 224 |
+
"learning_rate": 1.5467980295566506e-05,
|
| 225 |
+
"loss": 0.0624,
|
| 226 |
+
"step": 230
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 8.0,
|
| 230 |
+
"eval_loss": 0.02808324061334133,
|
| 231 |
+
"eval_runtime": 0.2003,
|
| 232 |
+
"eval_samples_per_second": 29.953,
|
| 233 |
+
"eval_steps_per_second": 9.984,
|
| 234 |
+
"step": 232
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 8.275862068965518,
|
| 238 |
+
"grad_norm": 1.7437331676483154,
|
| 239 |
+
"learning_rate": 1.5270935960591133e-05,
|
| 240 |
+
"loss": 0.0369,
|
| 241 |
+
"step": 240
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 8.620689655172415,
|
| 245 |
+
"grad_norm": 0.5273000597953796,
|
| 246 |
+
"learning_rate": 1.5073891625615764e-05,
|
| 247 |
+
"loss": 0.0499,
|
| 248 |
+
"step": 250
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 8.96551724137931,
|
| 252 |
+
"grad_norm": 0.6120426058769226,
|
| 253 |
+
"learning_rate": 1.4876847290640396e-05,
|
| 254 |
+
"loss": 0.0425,
|
| 255 |
+
"step": 260
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 9.0,
|
| 259 |
+
"eval_loss": 0.03035571426153183,
|
| 260 |
+
"eval_runtime": 0.1888,
|
| 261 |
+
"eval_samples_per_second": 31.778,
|
| 262 |
+
"eval_steps_per_second": 10.593,
|
| 263 |
+
"step": 261
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 9.310344827586206,
|
| 267 |
+
"grad_norm": 1.587663173675537,
|
| 268 |
+
"learning_rate": 1.4679802955665026e-05,
|
| 269 |
+
"loss": 0.0395,
|
| 270 |
+
"step": 270
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 9.655172413793103,
|
| 274 |
+
"grad_norm": 0.7260332703590393,
|
| 275 |
+
"learning_rate": 1.4482758620689657e-05,
|
| 276 |
+
"loss": 0.0493,
|
| 277 |
+
"step": 280
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 10.0,
|
| 281 |
+
"grad_norm": 0.9717508554458618,
|
| 282 |
+
"learning_rate": 1.4285714285714287e-05,
|
| 283 |
+
"loss": 0.0424,
|
| 284 |
+
"step": 290
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 10.0,
|
| 288 |
+
"eval_loss": 0.026125147938728333,
|
| 289 |
+
"eval_runtime": 0.0989,
|
| 290 |
+
"eval_samples_per_second": 60.684,
|
| 291 |
+
"eval_steps_per_second": 20.228,
|
| 292 |
+
"step": 290
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 10.344827586206897,
|
| 296 |
+
"grad_norm": 0.7487574815750122,
|
| 297 |
+
"learning_rate": 1.4088669950738918e-05,
|
| 298 |
+
"loss": 0.0412,
|
| 299 |
+
"step": 300
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 10.689655172413794,
|
| 303 |
+
"grad_norm": 1.3717066049575806,
|
| 304 |
+
"learning_rate": 1.3891625615763548e-05,
|
| 305 |
+
"loss": 0.0296,
|
| 306 |
+
"step": 310
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 11.0,
|
| 310 |
+
"eval_loss": 0.031011082231998444,
|
| 311 |
+
"eval_runtime": 0.1922,
|
| 312 |
+
"eval_samples_per_second": 31.22,
|
| 313 |
+
"eval_steps_per_second": 10.407,
|
| 314 |
+
"step": 319
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 11.03448275862069,
|
| 318 |
+
"grad_norm": 0.6860368847846985,
|
| 319 |
+
"learning_rate": 1.369458128078818e-05,
|
| 320 |
+
"loss": 0.0411,
|
| 321 |
+
"step": 320
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 11.379310344827585,
|
| 325 |
+
"grad_norm": 0.9999271035194397,
|
| 326 |
+
"learning_rate": 1.3497536945812807e-05,
|
| 327 |
+
"loss": 0.0358,
|
| 328 |
+
"step": 330
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 11.724137931034482,
|
| 332 |
+
"grad_norm": 1.2313721179962158,
|
| 333 |
+
"learning_rate": 1.330049261083744e-05,
|
| 334 |
+
"loss": 0.0324,
|
| 335 |
+
"step": 340
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 12.0,
|
| 339 |
+
"eval_loss": 0.029439905658364296,
|
| 340 |
+
"eval_runtime": 0.2015,
|
| 341 |
+
"eval_samples_per_second": 29.784,
|
| 342 |
+
"eval_steps_per_second": 9.928,
|
| 343 |
+
"step": 348
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 12.068965517241379,
|
| 347 |
+
"grad_norm": 0.9032502174377441,
|
| 348 |
+
"learning_rate": 1.310344827586207e-05,
|
| 349 |
+
"loss": 0.041,
|
| 350 |
+
"step": 350
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 12.413793103448276,
|
| 354 |
+
"grad_norm": 0.612800657749176,
|
| 355 |
+
"learning_rate": 1.29064039408867e-05,
|
| 356 |
+
"loss": 0.0384,
|
| 357 |
+
"step": 360
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 12.758620689655173,
|
| 361 |
+
"grad_norm": 0.19512540102005005,
|
| 362 |
+
"learning_rate": 1.2709359605911331e-05,
|
| 363 |
+
"loss": 0.0349,
|
| 364 |
+
"step": 370
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 13.0,
|
| 368 |
+
"eval_loss": 0.02956104278564453,
|
| 369 |
+
"eval_runtime": 0.1909,
|
| 370 |
+
"eval_samples_per_second": 31.424,
|
| 371 |
+
"eval_steps_per_second": 10.475,
|
| 372 |
+
"step": 377
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 13.10344827586207,
|
| 376 |
+
"grad_norm": 0.8481155633926392,
|
| 377 |
+
"learning_rate": 1.2512315270935961e-05,
|
| 378 |
+
"loss": 0.0271,
|
| 379 |
+
"step": 380
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 13.448275862068966,
|
| 383 |
+
"grad_norm": 1.3683249950408936,
|
| 384 |
+
"learning_rate": 1.2315270935960592e-05,
|
| 385 |
+
"loss": 0.0265,
|
| 386 |
+
"step": 390
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 13.793103448275861,
|
| 390 |
+
"grad_norm": 0.6365839838981628,
|
| 391 |
+
"learning_rate": 1.2118226600985224e-05,
|
| 392 |
+
"loss": 0.0298,
|
| 393 |
+
"step": 400
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 14.0,
|
| 397 |
+
"eval_loss": 0.03038203716278076,
|
| 398 |
+
"eval_runtime": 0.1902,
|
| 399 |
+
"eval_samples_per_second": 31.538,
|
| 400 |
+
"eval_steps_per_second": 10.513,
|
| 401 |
+
"step": 406
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 14.137931034482758,
|
| 405 |
+
"grad_norm": 0.7672634124755859,
|
| 406 |
+
"learning_rate": 1.1921182266009855e-05,
|
| 407 |
+
"loss": 0.0335,
|
| 408 |
+
"step": 410
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 14.482758620689655,
|
| 412 |
+
"grad_norm": 0.2541676163673401,
|
| 413 |
+
"learning_rate": 1.1724137931034483e-05,
|
| 414 |
+
"loss": 0.0286,
|
| 415 |
+
"step": 420
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 14.827586206896552,
|
| 419 |
+
"grad_norm": 0.8434980511665344,
|
| 420 |
+
"learning_rate": 1.1527093596059114e-05,
|
| 421 |
+
"loss": 0.0205,
|
| 422 |
+
"step": 430
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 15.0,
|
| 426 |
+
"eval_loss": 0.030394822359085083,
|
| 427 |
+
"eval_runtime": 0.19,
|
| 428 |
+
"eval_samples_per_second": 31.587,
|
| 429 |
+
"eval_steps_per_second": 10.529,
|
| 430 |
+
"step": 435
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 15.172413793103448,
|
| 434 |
+
"grad_norm": 0.4303562641143799,
|
| 435 |
+
"learning_rate": 1.1330049261083744e-05,
|
| 436 |
+
"loss": 0.0323,
|
| 437 |
+
"step": 440
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 15.517241379310345,
|
| 441 |
+
"grad_norm": 0.42710408568382263,
|
| 442 |
+
"learning_rate": 1.1133004926108375e-05,
|
| 443 |
+
"loss": 0.0227,
|
| 444 |
+
"step": 450
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 15.862068965517242,
|
| 448 |
+
"grad_norm": 0.6126664876937866,
|
| 449 |
+
"learning_rate": 1.0935960591133005e-05,
|
| 450 |
+
"loss": 0.0215,
|
| 451 |
+
"step": 460
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 16.0,
|
| 455 |
+
"eval_loss": 0.030341001227498055,
|
| 456 |
+
"eval_runtime": 0.1895,
|
| 457 |
+
"eval_samples_per_second": 31.669,
|
| 458 |
+
"eval_steps_per_second": 10.556,
|
| 459 |
+
"step": 464
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 16.20689655172414,
|
| 463 |
+
"grad_norm": 0.7011211514472961,
|
| 464 |
+
"learning_rate": 1.0738916256157637e-05,
|
| 465 |
+
"loss": 0.0337,
|
| 466 |
+
"step": 470
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 16.551724137931036,
|
| 470 |
+
"grad_norm": 0.5126092433929443,
|
| 471 |
+
"learning_rate": 1.0541871921182268e-05,
|
| 472 |
+
"loss": 0.0233,
|
| 473 |
+
"step": 480
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 16.896551724137932,
|
| 477 |
+
"grad_norm": 0.9033933281898499,
|
| 478 |
+
"learning_rate": 1.0344827586206898e-05,
|
| 479 |
+
"loss": 0.0182,
|
| 480 |
+
"step": 490
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 17.0,
|
| 484 |
+
"eval_loss": 0.027958964928984642,
|
| 485 |
+
"eval_runtime": 0.1902,
|
| 486 |
+
"eval_samples_per_second": 31.553,
|
| 487 |
+
"eval_steps_per_second": 10.518,
|
| 488 |
+
"step": 493
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 17.24137931034483,
|
| 492 |
+
"grad_norm": 0.4326847493648529,
|
| 493 |
+
"learning_rate": 1.0147783251231529e-05,
|
| 494 |
+
"loss": 0.0249,
|
| 495 |
+
"step": 500
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 17.586206896551722,
|
| 499 |
+
"grad_norm": 0.4034535586833954,
|
| 500 |
+
"learning_rate": 9.95073891625616e-06,
|
| 501 |
+
"loss": 0.0215,
|
| 502 |
+
"step": 510
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 17.93103448275862,
|
| 506 |
+
"grad_norm": 0.4999659061431885,
|
| 507 |
+
"learning_rate": 9.75369458128079e-06,
|
| 508 |
+
"loss": 0.0247,
|
| 509 |
+
"step": 520
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 18.0,
|
| 513 |
+
"eval_loss": 0.030091799795627594,
|
| 514 |
+
"eval_runtime": 0.1897,
|
| 515 |
+
"eval_samples_per_second": 31.624,
|
| 516 |
+
"eval_steps_per_second": 10.541,
|
| 517 |
+
"step": 522
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 18.275862068965516,
|
| 521 |
+
"grad_norm": 0.3004280924797058,
|
| 522 |
+
"learning_rate": 9.55665024630542e-06,
|
| 523 |
+
"loss": 0.0161,
|
| 524 |
+
"step": 530
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 18.620689655172413,
|
| 528 |
+
"grad_norm": 0.5018593072891235,
|
| 529 |
+
"learning_rate": 9.359605911330049e-06,
|
| 530 |
+
"loss": 0.0227,
|
| 531 |
+
"step": 540
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 18.96551724137931,
|
| 535 |
+
"grad_norm": 0.5728505849838257,
|
| 536 |
+
"learning_rate": 9.162561576354681e-06,
|
| 537 |
+
"loss": 0.0221,
|
| 538 |
+
"step": 550
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 19.0,
|
| 542 |
+
"eval_loss": 0.030765995383262634,
|
| 543 |
+
"eval_runtime": 0.1882,
|
| 544 |
+
"eval_samples_per_second": 31.886,
|
| 545 |
+
"eval_steps_per_second": 10.629,
|
| 546 |
+
"step": 551
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 19.310344827586206,
|
| 550 |
+
"grad_norm": 1.1493581533432007,
|
| 551 |
+
"learning_rate": 8.965517241379312e-06,
|
| 552 |
+
"loss": 0.0204,
|
| 553 |
+
"step": 560
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 19.655172413793103,
|
| 557 |
+
"grad_norm": 0.3401670753955841,
|
| 558 |
+
"learning_rate": 8.768472906403942e-06,
|
| 559 |
+
"loss": 0.0164,
|
| 560 |
+
"step": 570
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 20.0,
|
| 564 |
+
"grad_norm": 0.9875850081443787,
|
| 565 |
+
"learning_rate": 8.571428571428571e-06,
|
| 566 |
+
"loss": 0.0213,
|
| 567 |
+
"step": 580
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 20.0,
|
| 571 |
+
"eval_loss": 0.028663409873843193,
|
| 572 |
+
"eval_runtime": 0.1008,
|
| 573 |
+
"eval_samples_per_second": 59.511,
|
| 574 |
+
"eval_steps_per_second": 19.837,
|
| 575 |
+
"step": 580
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 20.344827586206897,
|
| 579 |
+
"grad_norm": 0.6725947260856628,
|
| 580 |
+
"learning_rate": 8.374384236453203e-06,
|
| 581 |
+
"loss": 0.0192,
|
| 582 |
+
"step": 590
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 20.689655172413794,
|
| 586 |
+
"grad_norm": 0.6594141721725464,
|
| 587 |
+
"learning_rate": 8.177339901477834e-06,
|
| 588 |
+
"loss": 0.0194,
|
| 589 |
+
"step": 600
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 21.0,
|
| 593 |
+
"eval_loss": 0.027452999725937843,
|
| 594 |
+
"eval_runtime": 0.1899,
|
| 595 |
+
"eval_samples_per_second": 31.602,
|
| 596 |
+
"eval_steps_per_second": 10.534,
|
| 597 |
+
"step": 609
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 21.03448275862069,
|
| 601 |
+
"grad_norm": 0.32469141483306885,
|
| 602 |
+
"learning_rate": 7.980295566502464e-06,
|
| 603 |
+
"loss": 0.0167,
|
| 604 |
+
"step": 610
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 21.379310344827587,
|
| 608 |
+
"grad_norm": 0.7029064893722534,
|
| 609 |
+
"learning_rate": 7.783251231527095e-06,
|
| 610 |
+
"loss": 0.0237,
|
| 611 |
+
"step": 620
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 21.724137931034484,
|
| 615 |
+
"grad_norm": 0.5634991526603699,
|
| 616 |
+
"learning_rate": 7.586206896551724e-06,
|
| 617 |
+
"loss": 0.0212,
|
| 618 |
+
"step": 630
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 22.0,
|
| 622 |
+
"eval_loss": 0.027833983302116394,
|
| 623 |
+
"eval_runtime": 0.1913,
|
| 624 |
+
"eval_samples_per_second": 31.357,
|
| 625 |
+
"eval_steps_per_second": 10.452,
|
| 626 |
+
"step": 638
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 22.06896551724138,
|
| 630 |
+
"grad_norm": 0.7171387672424316,
|
| 631 |
+
"learning_rate": 7.3891625615763555e-06,
|
| 632 |
+
"loss": 0.0204,
|
| 633 |
+
"step": 640
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 22.413793103448278,
|
| 637 |
+
"grad_norm": 0.3534681797027588,
|
| 638 |
+
"learning_rate": 7.192118226600986e-06,
|
| 639 |
+
"loss": 0.0162,
|
| 640 |
+
"step": 650
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 22.75862068965517,
|
| 644 |
+
"grad_norm": 0.2446085512638092,
|
| 645 |
+
"learning_rate": 6.995073891625616e-06,
|
| 646 |
+
"loss": 0.0178,
|
| 647 |
+
"step": 660
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 23.0,
|
| 651 |
+
"eval_loss": 0.0313410721719265,
|
| 652 |
+
"eval_runtime": 0.1923,
|
| 653 |
+
"eval_samples_per_second": 31.2,
|
| 654 |
+
"eval_steps_per_second": 10.4,
|
| 655 |
+
"step": 667
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 23.103448275862068,
|
| 659 |
+
"grad_norm": 0.3782992959022522,
|
| 660 |
+
"learning_rate": 6.798029556650246e-06,
|
| 661 |
+
"loss": 0.0214,
|
| 662 |
+
"step": 670
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 23.448275862068964,
|
| 666 |
+
"grad_norm": 0.2016129344701767,
|
| 667 |
+
"learning_rate": 6.600985221674877e-06,
|
| 668 |
+
"loss": 0.0167,
|
| 669 |
+
"step": 680
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 23.79310344827586,
|
| 673 |
+
"grad_norm": 0.3993048667907715,
|
| 674 |
+
"learning_rate": 6.403940886699508e-06,
|
| 675 |
+
"loss": 0.0217,
|
| 676 |
+
"step": 690
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 24.0,
|
| 680 |
+
"eval_loss": 0.030709436163306236,
|
| 681 |
+
"eval_runtime": 0.1914,
|
| 682 |
+
"eval_samples_per_second": 31.345,
|
| 683 |
+
"eval_steps_per_second": 10.448,
|
| 684 |
+
"step": 696
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 24.137931034482758,
|
| 688 |
+
"grad_norm": 0.7120524048805237,
|
| 689 |
+
"learning_rate": 6.206896551724138e-06,
|
| 690 |
+
"loss": 0.0227,
|
| 691 |
+
"step": 700
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 24.482758620689655,
|
| 695 |
+
"grad_norm": 0.3477053940296173,
|
| 696 |
+
"learning_rate": 6.00985221674877e-06,
|
| 697 |
+
"loss": 0.0203,
|
| 698 |
+
"step": 710
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 24.82758620689655,
|
| 702 |
+
"grad_norm": 0.30051225423812866,
|
| 703 |
+
"learning_rate": 5.812807881773399e-06,
|
| 704 |
+
"loss": 0.0118,
|
| 705 |
+
"step": 720
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 25.0,
|
| 709 |
+
"eval_loss": 0.02974347025156021,
|
| 710 |
+
"eval_runtime": 0.1936,
|
| 711 |
+
"eval_samples_per_second": 30.992,
|
| 712 |
+
"eval_steps_per_second": 10.331,
|
| 713 |
+
"step": 725
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 25.17241379310345,
|
| 717 |
+
"grad_norm": 0.38762152194976807,
|
| 718 |
+
"learning_rate": 5.61576354679803e-06,
|
| 719 |
+
"loss": 0.0161,
|
| 720 |
+
"step": 730
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 25.517241379310345,
|
| 724 |
+
"grad_norm": 0.48786938190460205,
|
| 725 |
+
"learning_rate": 5.41871921182266e-06,
|
| 726 |
+
"loss": 0.0153,
|
| 727 |
+
"step": 740
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 25.862068965517242,
|
| 731 |
+
"grad_norm": 0.273346483707428,
|
| 732 |
+
"learning_rate": 5.2216748768472915e-06,
|
| 733 |
+
"loss": 0.0189,
|
| 734 |
+
"step": 750
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 26.0,
|
| 738 |
+
"eval_loss": 0.031065121293067932,
|
| 739 |
+
"eval_runtime": 0.1912,
|
| 740 |
+
"eval_samples_per_second": 31.387,
|
| 741 |
+
"eval_steps_per_second": 10.462,
|
| 742 |
+
"step": 754
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 26.20689655172414,
|
| 746 |
+
"grad_norm": 0.8301162123680115,
|
| 747 |
+
"learning_rate": 5.024630541871922e-06,
|
| 748 |
+
"loss": 0.0194,
|
| 749 |
+
"step": 760
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 26.551724137931036,
|
| 753 |
+
"grad_norm": 0.46895724534988403,
|
| 754 |
+
"learning_rate": 4.8275862068965525e-06,
|
| 755 |
+
"loss": 0.0149,
|
| 756 |
+
"step": 770
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 26.896551724137932,
|
| 760 |
+
"grad_norm": 0.4784580171108246,
|
| 761 |
+
"learning_rate": 4.630541871921182e-06,
|
| 762 |
+
"loss": 0.0185,
|
| 763 |
+
"step": 780
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 27.0,
|
| 767 |
+
"eval_loss": 0.029767701402306557,
|
| 768 |
+
"eval_runtime": 0.1921,
|
| 769 |
+
"eval_samples_per_second": 31.24,
|
| 770 |
+
"eval_steps_per_second": 10.413,
|
| 771 |
+
"step": 783
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 27.24137931034483,
|
| 775 |
+
"grad_norm": 0.7427147626876831,
|
| 776 |
+
"learning_rate": 4.4334975369458135e-06,
|
| 777 |
+
"loss": 0.0155,
|
| 778 |
+
"step": 790
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 27.586206896551722,
|
| 782 |
+
"grad_norm": 0.4144653081893921,
|
| 783 |
+
"learning_rate": 4.236453201970444e-06,
|
| 784 |
+
"loss": 0.0159,
|
| 785 |
+
"step": 800
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 27.93103448275862,
|
| 789 |
+
"grad_norm": 0.29942747950553894,
|
| 790 |
+
"learning_rate": 4.039408866995074e-06,
|
| 791 |
+
"loss": 0.021,
|
| 792 |
+
"step": 810
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 28.0,
|
| 796 |
+
"eval_loss": 0.029253564774990082,
|
| 797 |
+
"eval_runtime": 0.1914,
|
| 798 |
+
"eval_samples_per_second": 31.352,
|
| 799 |
+
"eval_steps_per_second": 10.451,
|
| 800 |
+
"step": 812
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 28.275862068965516,
|
| 804 |
+
"grad_norm": 0.21119730174541473,
|
| 805 |
+
"learning_rate": 3.842364532019705e-06,
|
| 806 |
+
"loss": 0.0153,
|
| 807 |
+
"step": 820
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 28.620689655172413,
|
| 811 |
+
"grad_norm": 0.2980373501777649,
|
| 812 |
+
"learning_rate": 3.6453201970443354e-06,
|
| 813 |
+
"loss": 0.0189,
|
| 814 |
+
"step": 830
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 28.96551724137931,
|
| 818 |
+
"grad_norm": 0.09113238751888275,
|
| 819 |
+
"learning_rate": 3.448275862068966e-06,
|
| 820 |
+
"loss": 0.0102,
|
| 821 |
+
"step": 840
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 29.0,
|
| 825 |
+
"eval_loss": 0.028439467772841454,
|
| 826 |
+
"eval_runtime": 0.1908,
|
| 827 |
+
"eval_samples_per_second": 31.449,
|
| 828 |
+
"eval_steps_per_second": 10.483,
|
| 829 |
+
"step": 841
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 29.310344827586206,
|
| 833 |
+
"grad_norm": 0.3905262351036072,
|
| 834 |
+
"learning_rate": 3.2512315270935963e-06,
|
| 835 |
+
"loss": 0.0153,
|
| 836 |
+
"step": 850
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 29.655172413793103,
|
| 840 |
+
"grad_norm": 0.36752381920814514,
|
| 841 |
+
"learning_rate": 3.054187192118227e-06,
|
| 842 |
+
"loss": 0.014,
|
| 843 |
+
"step": 860
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 30.0,
|
| 847 |
+
"grad_norm": 0.6769506931304932,
|
| 848 |
+
"learning_rate": 2.8571428571428573e-06,
|
| 849 |
+
"loss": 0.018,
|
| 850 |
+
"step": 870
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 30.0,
|
| 854 |
+
"eval_loss": 0.02924039028584957,
|
| 855 |
+
"eval_runtime": 0.102,
|
| 856 |
+
"eval_samples_per_second": 58.797,
|
| 857 |
+
"eval_steps_per_second": 19.599,
|
| 858 |
+
"step": 870
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 30.344827586206897,
|
| 862 |
+
"grad_norm": 0.676186740398407,
|
| 863 |
+
"learning_rate": 2.660098522167488e-06,
|
| 864 |
+
"loss": 0.0202,
|
| 865 |
+
"step": 880
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 30.689655172413794,
|
| 869 |
+
"grad_norm": 0.2909271717071533,
|
| 870 |
+
"learning_rate": 2.4630541871921186e-06,
|
| 871 |
+
"loss": 0.0105,
|
| 872 |
+
"step": 890
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 31.0,
|
| 876 |
+
"eval_loss": 0.030303308740258217,
|
| 877 |
+
"eval_runtime": 0.191,
|
| 878 |
+
"eval_samples_per_second": 31.415,
|
| 879 |
+
"eval_steps_per_second": 10.472,
|
| 880 |
+
"step": 899
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 31.03448275862069,
|
| 884 |
+
"grad_norm": 0.8155515193939209,
|
| 885 |
+
"learning_rate": 2.266009852216749e-06,
|
| 886 |
+
"loss": 0.0174,
|
| 887 |
+
"step": 900
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 31.379310344827587,
|
| 891 |
+
"grad_norm": 0.5190662741661072,
|
| 892 |
+
"learning_rate": 2.0689655172413796e-06,
|
| 893 |
+
"loss": 0.0157,
|
| 894 |
+
"step": 910
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 31.724137931034484,
|
| 898 |
+
"grad_norm": 0.8997210264205933,
|
| 899 |
+
"learning_rate": 1.8719211822660098e-06,
|
| 900 |
+
"loss": 0.0155,
|
| 901 |
+
"step": 920
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 32.0,
|
| 905 |
+
"eval_loss": 0.030399195849895477,
|
| 906 |
+
"eval_runtime": 0.1899,
|
| 907 |
+
"eval_samples_per_second": 31.593,
|
| 908 |
+
"eval_steps_per_second": 10.531,
|
| 909 |
+
"step": 928
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 32.06896551724138,
|
| 913 |
+
"grad_norm": 0.5701755881309509,
|
| 914 |
+
"learning_rate": 1.6748768472906405e-06,
|
| 915 |
+
"loss": 0.0139,
|
| 916 |
+
"step": 930
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 32.41379310344828,
|
| 920 |
+
"grad_norm": 0.16824859380722046,
|
| 921 |
+
"learning_rate": 1.4778325123152712e-06,
|
| 922 |
+
"loss": 0.0113,
|
| 923 |
+
"step": 940
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 32.758620689655174,
|
| 927 |
+
"grad_norm": 0.5066978931427002,
|
| 928 |
+
"learning_rate": 1.2807881773399017e-06,
|
| 929 |
+
"loss": 0.0127,
|
| 930 |
+
"step": 950
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 33.0,
|
| 934 |
+
"eval_loss": 0.030721982941031456,
|
| 935 |
+
"eval_runtime": 0.191,
|
| 936 |
+
"eval_samples_per_second": 31.41,
|
| 937 |
+
"eval_steps_per_second": 10.47,
|
| 938 |
+
"step": 957
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 33.10344827586207,
|
| 942 |
+
"grad_norm": 0.5038828253746033,
|
| 943 |
+
"learning_rate": 1.0837438423645322e-06,
|
| 944 |
+
"loss": 0.0138,
|
| 945 |
+
"step": 960
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 33.44827586206897,
|
| 949 |
+
"grad_norm": 0.4406326115131378,
|
| 950 |
+
"learning_rate": 8.866995073891626e-07,
|
| 951 |
+
"loss": 0.0116,
|
| 952 |
+
"step": 970
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 33.793103448275865,
|
| 956 |
+
"grad_norm": 0.3732337951660156,
|
| 957 |
+
"learning_rate": 6.896551724137931e-07,
|
| 958 |
+
"loss": 0.0171,
|
| 959 |
+
"step": 980
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 34.0,
|
| 963 |
+
"eval_loss": 0.03042110800743103,
|
| 964 |
+
"eval_runtime": 0.191,
|
| 965 |
+
"eval_samples_per_second": 31.415,
|
| 966 |
+
"eval_steps_per_second": 10.472,
|
| 967 |
+
"step": 986
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 34.13793103448276,
|
| 971 |
+
"grad_norm": 0.32739463448524475,
|
| 972 |
+
"learning_rate": 4.926108374384237e-07,
|
| 973 |
+
"loss": 0.0186,
|
| 974 |
+
"step": 990
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 34.48275862068966,
|
| 978 |
+
"grad_norm": 0.45314541459083557,
|
| 979 |
+
"learning_rate": 2.955665024630542e-07,
|
| 980 |
+
"loss": 0.0108,
|
| 981 |
+
"step": 1000
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 34.827586206896555,
|
| 985 |
+
"grad_norm": 0.8981533050537109,
|
| 986 |
+
"learning_rate": 9.852216748768474e-08,
|
| 987 |
+
"loss": 0.0157,
|
| 988 |
+
"step": 1010
|
| 989 |
+
}
|
| 990 |
+
],
|
| 991 |
+
"logging_steps": 10,
|
| 992 |
+
"max_steps": 1015,
|
| 993 |
+
"num_input_tokens_seen": 0,
|
| 994 |
+
"num_train_epochs": 35,
|
| 995 |
+
"save_steps": 500,
|
| 996 |
+
"stateful_callbacks": {
|
| 997 |
+
"TrainerControl": {
|
| 998 |
+
"args": {
|
| 999 |
+
"should_epoch_stop": false,
|
| 1000 |
+
"should_evaluate": false,
|
| 1001 |
+
"should_log": false,
|
| 1002 |
+
"should_save": true,
|
| 1003 |
+
"should_training_stop": true
|
| 1004 |
+
},
|
| 1005 |
+
"attributes": {}
|
| 1006 |
+
}
|
| 1007 |
+
},
|
| 1008 |
+
"total_flos": 2159634677760000.0,
|
| 1009 |
+
"train_batch_size": 4,
|
| 1010 |
+
"trial_name": null,
|
| 1011 |
+
"trial_params": null
|
| 1012 |
+
}
|
json_extraction_point_activity/checkpoint-1015/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:333cc851b675d7a01620eeba59cdfbc1a624d74927d086ab660bd04fef9b6029
|
| 3 |
+
size 5240
|
json_extraction_point_activity/checkpoint-500/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2950734544
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:87363b3a1b2a51942b6c278f69df9a806706e5d83c60c65870a507e043356124
|
| 3 |
size 2950734544
|
json_extraction_point_activity/checkpoint-500/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5901778825
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:26c2d8aae51544bc1ae1adf721bf158fe114281c194c7e88ada3907e4afe56eb
|
| 3 |
size 5901778825
|
json_extraction_point_activity/checkpoint-500/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2533398c3a2262d5247b0295eea340c12039f2768e0c6be8c67c59019ab0553c
|
| 3 |
size 14244
|
json_extraction_point_activity/checkpoint-500/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fac53bd556f9070ef9763eb8ffdc7241c2002524738db57098477c02123b9841
|
| 3 |
size 1064
|
json_extraction_point_activity/checkpoint-500/trainer_state.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
"global_step": 500,
|
| 7 |
"is_hyper_param_search": false,
|
|
@@ -9,518 +9,494 @@
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate": 1.
|
| 15 |
-
"loss":
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate": 1.
|
| 22 |
-
"loss":
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 1.0,
|
| 27 |
-
"eval_loss":
|
| 28 |
-
"eval_runtime": 0.
|
| 29 |
-
"eval_samples_per_second":
|
| 30 |
-
"eval_steps_per_second": 10.
|
| 31 |
-
"step":
|
| 32 |
},
|
| 33 |
{
|
| 34 |
-
"epoch": 1.
|
| 35 |
-
"grad_norm":
|
| 36 |
-
"learning_rate": 1.
|
| 37 |
-
"loss":
|
| 38 |
"step": 30
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch": 1.
|
| 42 |
-
"grad_norm": 5.
|
| 43 |
-
"learning_rate": 1.
|
| 44 |
-
"loss": 1.
|
| 45 |
"step": 40
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch":
|
| 49 |
-
"
|
| 50 |
-
"
|
| 51 |
-
"
|
| 52 |
-
"
|
| 53 |
-
"step": 48
|
| 54 |
},
|
| 55 |
{
|
| 56 |
-
"epoch": 2.
|
| 57 |
-
"
|
| 58 |
-
"
|
| 59 |
-
"
|
| 60 |
-
"
|
|
|
|
| 61 |
},
|
| 62 |
{
|
| 63 |
-
"epoch": 2.
|
| 64 |
-
"grad_norm":
|
| 65 |
-
"learning_rate": 1.
|
| 66 |
-
"loss": 0.
|
| 67 |
"step": 60
|
| 68 |
},
|
| 69 |
{
|
| 70 |
-
"epoch": 2.
|
| 71 |
-
"grad_norm": 2.
|
| 72 |
-
"learning_rate": 1.
|
| 73 |
-
"loss": 0.
|
| 74 |
"step": 70
|
| 75 |
},
|
| 76 |
{
|
| 77 |
-
"epoch":
|
| 78 |
-
"
|
| 79 |
-
"
|
| 80 |
-
"
|
| 81 |
-
"eval_steps_per_second": 10.36,
|
| 82 |
-
"step": 72
|
| 83 |
-
},
|
| 84 |
-
{
|
| 85 |
-
"epoch": 3.3333333333333335,
|
| 86 |
-
"grad_norm": 1.314430832862854,
|
| 87 |
-
"learning_rate": 1.8095238095238097e-05,
|
| 88 |
-
"loss": 0.2508,
|
| 89 |
"step": 80
|
| 90 |
},
|
| 91 |
{
|
| 92 |
-
"epoch": 3.
|
| 93 |
-
"
|
| 94 |
-
"
|
| 95 |
-
"
|
| 96 |
-
"
|
|
|
|
| 97 |
},
|
| 98 |
{
|
| 99 |
-
"epoch":
|
| 100 |
-
"
|
| 101 |
-
"
|
| 102 |
-
"
|
| 103 |
-
"
|
| 104 |
-
"step": 96
|
| 105 |
},
|
| 106 |
{
|
| 107 |
-
"epoch":
|
| 108 |
-
"grad_norm": 1.
|
| 109 |
-
"learning_rate": 1.
|
| 110 |
-
"loss": 0.
|
| 111 |
"step": 100
|
| 112 |
},
|
| 113 |
{
|
| 114 |
-
"epoch":
|
| 115 |
-
"grad_norm": 1.
|
| 116 |
-
"learning_rate": 1.
|
| 117 |
-
"loss": 0.
|
| 118 |
"step": 110
|
| 119 |
},
|
| 120 |
{
|
| 121 |
-
"epoch":
|
| 122 |
-
"
|
| 123 |
-
"
|
| 124 |
-
"
|
| 125 |
-
"
|
|
|
|
| 126 |
},
|
| 127 |
{
|
| 128 |
-
"epoch":
|
| 129 |
-
"
|
| 130 |
-
"
|
| 131 |
-
"
|
| 132 |
-
"eval_steps_per_second": 20.192,
|
| 133 |
"step": 120
|
| 134 |
},
|
| 135 |
{
|
| 136 |
-
"epoch":
|
| 137 |
-
"grad_norm":
|
| 138 |
-
"learning_rate": 1.
|
| 139 |
-
"loss": 0.
|
| 140 |
"step": 130
|
| 141 |
},
|
| 142 |
{
|
| 143 |
-
"epoch":
|
| 144 |
-
"grad_norm":
|
| 145 |
-
"learning_rate": 1.
|
| 146 |
-
"loss": 0.
|
| 147 |
"step": 140
|
| 148 |
},
|
| 149 |
{
|
| 150 |
-
"epoch":
|
| 151 |
-
"eval_loss": 0.
|
| 152 |
-
"eval_runtime": 0.
|
| 153 |
-
"eval_samples_per_second":
|
| 154 |
-
"eval_steps_per_second": 10.
|
| 155 |
-
"step":
|
| 156 |
},
|
| 157 |
{
|
| 158 |
-
"epoch":
|
| 159 |
-
"grad_norm": 0.
|
| 160 |
-
"learning_rate": 1.
|
| 161 |
-
"loss": 0.
|
| 162 |
"step": 150
|
| 163 |
},
|
| 164 |
{
|
| 165 |
-
"epoch":
|
| 166 |
-
"grad_norm":
|
| 167 |
-
"learning_rate": 1.
|
| 168 |
-
"loss": 0.
|
| 169 |
"step": 160
|
| 170 |
},
|
| 171 |
{
|
| 172 |
-
"epoch":
|
| 173 |
-
"
|
| 174 |
-
"
|
| 175 |
-
"
|
| 176 |
-
"
|
| 177 |
-
"step": 168
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch":
|
| 181 |
-
"
|
| 182 |
-
"
|
| 183 |
-
"
|
| 184 |
-
"
|
|
|
|
| 185 |
},
|
| 186 |
{
|
| 187 |
-
"epoch":
|
| 188 |
-
"grad_norm": 0.
|
| 189 |
-
"learning_rate": 1.
|
| 190 |
-
"loss": 0.
|
| 191 |
"step": 180
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch":
|
| 195 |
-
"grad_norm":
|
| 196 |
-
"learning_rate": 1.
|
| 197 |
-
"loss": 0.
|
| 198 |
"step": 190
|
| 199 |
},
|
| 200 |
{
|
| 201 |
-
"epoch":
|
| 202 |
-
"
|
| 203 |
-
"
|
| 204 |
-
"
|
| 205 |
-
"eval_steps_per_second": 10.655,
|
| 206 |
-
"step": 192
|
| 207 |
-
},
|
| 208 |
-
{
|
| 209 |
-
"epoch": 8.333333333333334,
|
| 210 |
-
"grad_norm": 1.0495927333831787,
|
| 211 |
-
"learning_rate": 1.523809523809524e-05,
|
| 212 |
-
"loss": 0.0474,
|
| 213 |
"step": 200
|
| 214 |
},
|
| 215 |
{
|
| 216 |
-
"epoch":
|
| 217 |
-
"
|
| 218 |
-
"
|
| 219 |
-
"
|
| 220 |
-
"
|
|
|
|
| 221 |
},
|
| 222 |
{
|
| 223 |
-
"epoch":
|
| 224 |
-
"
|
| 225 |
-
"
|
| 226 |
-
"
|
| 227 |
-
"
|
| 228 |
-
"step": 216
|
| 229 |
},
|
| 230 |
{
|
| 231 |
-
"epoch":
|
| 232 |
-
"grad_norm":
|
| 233 |
-
"learning_rate": 1.
|
| 234 |
-
"loss": 0.
|
| 235 |
"step": 220
|
| 236 |
},
|
| 237 |
{
|
| 238 |
-
"epoch":
|
| 239 |
-
"grad_norm": 0.
|
| 240 |
-
"learning_rate": 1.
|
| 241 |
-
"loss": 0.
|
| 242 |
"step": 230
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch":
|
| 246 |
-
"
|
| 247 |
-
"
|
| 248 |
-
"
|
| 249 |
-
"
|
|
|
|
| 250 |
},
|
| 251 |
{
|
| 252 |
-
"epoch":
|
| 253 |
-
"
|
| 254 |
-
"
|
| 255 |
-
"
|
| 256 |
-
"eval_steps_per_second": 20.311,
|
| 257 |
"step": 240
|
| 258 |
},
|
| 259 |
{
|
| 260 |
-
"epoch":
|
| 261 |
-
"grad_norm": 0.
|
| 262 |
-
"learning_rate": 1.
|
| 263 |
-
"loss": 0.
|
| 264 |
"step": 250
|
| 265 |
},
|
| 266 |
{
|
| 267 |
-
"epoch":
|
| 268 |
-
"grad_norm": 0.
|
| 269 |
-
"learning_rate": 1.
|
| 270 |
-
"loss": 0.
|
| 271 |
"step": 260
|
| 272 |
},
|
| 273 |
{
|
| 274 |
-
"epoch":
|
| 275 |
-
"eval_loss": 0.
|
| 276 |
-
"eval_runtime": 0.
|
| 277 |
-
"eval_samples_per_second":
|
| 278 |
-
"eval_steps_per_second": 10.
|
| 279 |
-
"step":
|
| 280 |
},
|
| 281 |
{
|
| 282 |
-
"epoch":
|
| 283 |
-
"grad_norm":
|
| 284 |
-
"learning_rate": 1.
|
| 285 |
-
"loss": 0.
|
| 286 |
"step": 270
|
| 287 |
},
|
| 288 |
{
|
| 289 |
-
"epoch":
|
| 290 |
-
"grad_norm":
|
| 291 |
-
"learning_rate": 1.
|
| 292 |
-
"loss": 0.
|
| 293 |
"step": 280
|
| 294 |
},
|
| 295 |
{
|
| 296 |
-
"epoch":
|
| 297 |
-
"
|
| 298 |
-
"
|
| 299 |
-
"
|
| 300 |
-
"
|
| 301 |
-
"step": 288
|
| 302 |
},
|
| 303 |
{
|
| 304 |
-
"epoch":
|
| 305 |
-
"
|
| 306 |
-
"
|
| 307 |
-
"
|
|
|
|
| 308 |
"step": 290
|
| 309 |
},
|
| 310 |
{
|
| 311 |
-
"epoch":
|
| 312 |
-
"grad_norm": 0.
|
| 313 |
-
"learning_rate": 1.
|
| 314 |
-
"loss": 0.
|
| 315 |
"step": 300
|
| 316 |
},
|
| 317 |
{
|
| 318 |
-
"epoch":
|
| 319 |
-
"grad_norm":
|
| 320 |
-
"learning_rate": 1.
|
| 321 |
-
"loss": 0.
|
| 322 |
"step": 310
|
| 323 |
},
|
| 324 |
{
|
| 325 |
-
"epoch":
|
| 326 |
-
"eval_loss": 0.
|
| 327 |
-
"eval_runtime": 0.
|
| 328 |
-
"eval_samples_per_second":
|
| 329 |
-
"eval_steps_per_second": 10.
|
| 330 |
-
"step":
|
| 331 |
},
|
| 332 |
{
|
| 333 |
-
"epoch":
|
| 334 |
-
"grad_norm": 0.
|
| 335 |
-
"learning_rate": 1.
|
| 336 |
-
"loss": 0.
|
| 337 |
"step": 320
|
| 338 |
},
|
| 339 |
{
|
| 340 |
-
"epoch":
|
| 341 |
-
"grad_norm": 0.
|
| 342 |
-
"learning_rate": 1.
|
| 343 |
-
"loss": 0.
|
| 344 |
"step": 330
|
| 345 |
},
|
| 346 |
{
|
| 347 |
-
"epoch":
|
| 348 |
-
"
|
| 349 |
-
"
|
| 350 |
-
"
|
| 351 |
-
"eval_steps_per_second": 10.271,
|
| 352 |
-
"step": 336
|
| 353 |
-
},
|
| 354 |
-
{
|
| 355 |
-
"epoch": 14.166666666666666,
|
| 356 |
-
"grad_norm": 0.6511815786361694,
|
| 357 |
-
"learning_rate": 1.1904761904761905e-05,
|
| 358 |
-
"loss": 0.0318,
|
| 359 |
"step": 340
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch":
|
| 363 |
-
"
|
| 364 |
-
"
|
| 365 |
-
"
|
| 366 |
-
"
|
|
|
|
| 367 |
},
|
| 368 |
{
|
| 369 |
-
"epoch":
|
| 370 |
-
"grad_norm": 0.
|
| 371 |
-
"learning_rate": 1.
|
| 372 |
-
"loss": 0.
|
| 373 |
-
"step":
|
| 374 |
},
|
| 375 |
{
|
| 376 |
-
"epoch":
|
| 377 |
-
"
|
| 378 |
-
"
|
| 379 |
-
"
|
| 380 |
-
"eval_steps_per_second": 19.977,
|
| 381 |
"step": 360
|
| 382 |
},
|
| 383 |
{
|
| 384 |
-
"epoch":
|
| 385 |
-
"grad_norm":
|
| 386 |
-
"learning_rate": 1.
|
| 387 |
-
"loss": 0.
|
| 388 |
"step": 370
|
| 389 |
},
|
| 390 |
{
|
| 391 |
-
"epoch":
|
| 392 |
-
"
|
| 393 |
-
"
|
| 394 |
-
"
|
| 395 |
-
"
|
|
|
|
| 396 |
},
|
| 397 |
{
|
| 398 |
-
"epoch":
|
| 399 |
-
"
|
| 400 |
-
"
|
| 401 |
-
"
|
| 402 |
-
"
|
| 403 |
-
"step": 384
|
| 404 |
},
|
| 405 |
{
|
| 406 |
-
"epoch":
|
| 407 |
-
"grad_norm":
|
| 408 |
-
"learning_rate": 1.
|
| 409 |
-
"loss": 0.
|
| 410 |
"step": 390
|
| 411 |
},
|
| 412 |
{
|
| 413 |
-
"epoch":
|
| 414 |
-
"grad_norm": 0.
|
| 415 |
-
"learning_rate": 1.
|
| 416 |
-
"loss": 0.
|
| 417 |
"step": 400
|
| 418 |
},
|
| 419 |
{
|
| 420 |
-
"epoch":
|
| 421 |
-
"eval_loss": 0.
|
| 422 |
-
"eval_runtime": 0.
|
| 423 |
-
"eval_samples_per_second":
|
| 424 |
-
"eval_steps_per_second": 10.
|
| 425 |
-
"step":
|
| 426 |
},
|
| 427 |
{
|
| 428 |
-
"epoch":
|
| 429 |
-
"grad_norm": 0.
|
| 430 |
-
"learning_rate": 1.
|
| 431 |
-
"loss": 0.
|
| 432 |
"step": 410
|
| 433 |
},
|
| 434 |
{
|
| 435 |
-
"epoch":
|
| 436 |
-
"grad_norm": 0.
|
| 437 |
-
"learning_rate":
|
| 438 |
-
"loss": 0.
|
| 439 |
"step": 420
|
| 440 |
},
|
| 441 |
{
|
| 442 |
-
"epoch":
|
| 443 |
-
"grad_norm": 0.
|
| 444 |
-
"learning_rate":
|
| 445 |
-
"loss": 0.
|
| 446 |
"step": 430
|
| 447 |
},
|
| 448 |
{
|
| 449 |
-
"epoch":
|
| 450 |
-
"eval_loss": 0.
|
| 451 |
-
"eval_runtime": 0.
|
| 452 |
-
"eval_samples_per_second":
|
| 453 |
-
"eval_steps_per_second": 10.
|
| 454 |
-
"step":
|
| 455 |
},
|
| 456 |
{
|
| 457 |
-
"epoch":
|
| 458 |
-
"grad_norm": 0.
|
| 459 |
-
"learning_rate":
|
| 460 |
-
"loss": 0.
|
| 461 |
"step": 440
|
| 462 |
},
|
| 463 |
{
|
| 464 |
-
"epoch":
|
| 465 |
-
"grad_norm": 0.
|
| 466 |
-
"learning_rate":
|
| 467 |
-
"loss": 0.
|
| 468 |
"step": 450
|
| 469 |
},
|
| 470 |
{
|
| 471 |
-
"epoch":
|
| 472 |
-
"
|
| 473 |
-
"
|
| 474 |
-
"
|
| 475 |
-
"
|
| 476 |
-
"step": 456
|
| 477 |
},
|
| 478 |
{
|
| 479 |
-
"epoch":
|
| 480 |
-
"
|
| 481 |
-
"
|
| 482 |
-
"
|
| 483 |
-
"
|
|
|
|
| 484 |
},
|
| 485 |
{
|
| 486 |
-
"epoch":
|
| 487 |
-
"grad_norm": 0.
|
| 488 |
-
"learning_rate":
|
| 489 |
-
"loss": 0.
|
| 490 |
"step": 470
|
| 491 |
},
|
| 492 |
{
|
| 493 |
-
"epoch":
|
| 494 |
-
"grad_norm": 0.
|
| 495 |
-
"learning_rate":
|
| 496 |
-
"loss": 0.
|
| 497 |
"step": 480
|
| 498 |
},
|
| 499 |
{
|
| 500 |
-
"epoch":
|
| 501 |
-
"
|
| 502 |
-
"
|
| 503 |
-
"
|
| 504 |
-
"
|
| 505 |
-
"step": 480
|
| 506 |
},
|
| 507 |
{
|
| 508 |
-
"epoch":
|
| 509 |
-
"
|
| 510 |
-
"
|
| 511 |
-
"
|
| 512 |
-
"
|
|
|
|
| 513 |
},
|
| 514 |
{
|
| 515 |
-
"epoch":
|
| 516 |
-
"grad_norm": 0.
|
| 517 |
-
"learning_rate":
|
| 518 |
-
"loss": 0.
|
| 519 |
"step": 500
|
| 520 |
}
|
| 521 |
],
|
| 522 |
"logging_steps": 10,
|
| 523 |
-
"max_steps":
|
| 524 |
"num_input_tokens_seen": 0,
|
| 525 |
"num_train_epochs": 35,
|
| 526 |
"save_steps": 500,
|
|
@@ -536,7 +512,7 @@
|
|
| 536 |
"attributes": {}
|
| 537 |
}
|
| 538 |
},
|
| 539 |
-
"total_flos":
|
| 540 |
"train_batch_size": 4,
|
| 541 |
"trial_name": null,
|
| 542 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 17.24137931034483,
|
| 5 |
"eval_steps": 500,
|
| 6 |
"global_step": 500,
|
| 7 |
"is_hyper_param_search": false,
|
|
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.3448275862068966,
|
| 13 |
+
"grad_norm": 174.2040557861328,
|
| 14 |
+
"learning_rate": 1.9802955665024632e-05,
|
| 15 |
+
"loss": 11.9232,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.6896551724137931,
|
| 20 |
+
"grad_norm": 54.92245101928711,
|
| 21 |
+
"learning_rate": 1.9605911330049263e-05,
|
| 22 |
+
"loss": 6.0635,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 1.0,
|
| 27 |
+
"eval_loss": 0.8173177242279053,
|
| 28 |
+
"eval_runtime": 0.1894,
|
| 29 |
+
"eval_samples_per_second": 31.679,
|
| 30 |
+
"eval_steps_per_second": 10.56,
|
| 31 |
+
"step": 29
|
| 32 |
},
|
| 33 |
{
|
| 34 |
+
"epoch": 1.0344827586206897,
|
| 35 |
+
"grad_norm": 16.176136016845703,
|
| 36 |
+
"learning_rate": 1.9408866995073893e-05,
|
| 37 |
+
"loss": 2.7637,
|
| 38 |
"step": 30
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 1.3793103448275863,
|
| 42 |
+
"grad_norm": 5.2406907081604,
|
| 43 |
+
"learning_rate": 1.9211822660098524e-05,
|
| 44 |
+
"loss": 1.0566,
|
| 45 |
"step": 40
|
| 46 |
},
|
| 47 |
{
|
| 48 |
+
"epoch": 1.7241379310344827,
|
| 49 |
+
"grad_norm": 3.6556057929992676,
|
| 50 |
+
"learning_rate": 1.9014778325123154e-05,
|
| 51 |
+
"loss": 0.6543,
|
| 52 |
+
"step": 50
|
|
|
|
| 53 |
},
|
| 54 |
{
|
| 55 |
+
"epoch": 2.0,
|
| 56 |
+
"eval_loss": 0.26999202370643616,
|
| 57 |
+
"eval_runtime": 0.1963,
|
| 58 |
+
"eval_samples_per_second": 30.571,
|
| 59 |
+
"eval_steps_per_second": 10.19,
|
| 60 |
+
"step": 58
|
| 61 |
},
|
| 62 |
{
|
| 63 |
+
"epoch": 2.0689655172413794,
|
| 64 |
+
"grad_norm": 1.8488504886627197,
|
| 65 |
+
"learning_rate": 1.8817733990147784e-05,
|
| 66 |
+
"loss": 0.4713,
|
| 67 |
"step": 60
|
| 68 |
},
|
| 69 |
{
|
| 70 |
+
"epoch": 2.413793103448276,
|
| 71 |
+
"grad_norm": 2.723362922668457,
|
| 72 |
+
"learning_rate": 1.8620689655172415e-05,
|
| 73 |
+
"loss": 0.3144,
|
| 74 |
"step": 70
|
| 75 |
},
|
| 76 |
{
|
| 77 |
+
"epoch": 2.7586206896551726,
|
| 78 |
+
"grad_norm": 1.8972634077072144,
|
| 79 |
+
"learning_rate": 1.8423645320197045e-05,
|
| 80 |
+
"loss": 0.2604,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 81 |
"step": 80
|
| 82 |
},
|
| 83 |
{
|
| 84 |
+
"epoch": 3.0,
|
| 85 |
+
"eval_loss": 0.09994816780090332,
|
| 86 |
+
"eval_runtime": 0.1912,
|
| 87 |
+
"eval_samples_per_second": 31.381,
|
| 88 |
+
"eval_steps_per_second": 10.46,
|
| 89 |
+
"step": 87
|
| 90 |
},
|
| 91 |
{
|
| 92 |
+
"epoch": 3.103448275862069,
|
| 93 |
+
"grad_norm": 3.115511417388916,
|
| 94 |
+
"learning_rate": 1.8226600985221676e-05,
|
| 95 |
+
"loss": 0.2067,
|
| 96 |
+
"step": 90
|
|
|
|
| 97 |
},
|
| 98 |
{
|
| 99 |
+
"epoch": 3.4482758620689653,
|
| 100 |
+
"grad_norm": 1.7388259172439575,
|
| 101 |
+
"learning_rate": 1.8029556650246306e-05,
|
| 102 |
+
"loss": 0.1494,
|
| 103 |
"step": 100
|
| 104 |
},
|
| 105 |
{
|
| 106 |
+
"epoch": 3.793103448275862,
|
| 107 |
+
"grad_norm": 1.2075275182724,
|
| 108 |
+
"learning_rate": 1.7832512315270937e-05,
|
| 109 |
+
"loss": 0.1411,
|
| 110 |
"step": 110
|
| 111 |
},
|
| 112 |
{
|
| 113 |
+
"epoch": 4.0,
|
| 114 |
+
"eval_loss": 0.05977391079068184,
|
| 115 |
+
"eval_runtime": 0.1933,
|
| 116 |
+
"eval_samples_per_second": 31.034,
|
| 117 |
+
"eval_steps_per_second": 10.345,
|
| 118 |
+
"step": 116
|
| 119 |
},
|
| 120 |
{
|
| 121 |
+
"epoch": 4.137931034482759,
|
| 122 |
+
"grad_norm": 1.7329133749008179,
|
| 123 |
+
"learning_rate": 1.7635467980295567e-05,
|
| 124 |
+
"loss": 0.1196,
|
|
|
|
| 125 |
"step": 120
|
| 126 |
},
|
| 127 |
{
|
| 128 |
+
"epoch": 4.482758620689655,
|
| 129 |
+
"grad_norm": 2.210278272628784,
|
| 130 |
+
"learning_rate": 1.7438423645320198e-05,
|
| 131 |
+
"loss": 0.1132,
|
| 132 |
"step": 130
|
| 133 |
},
|
| 134 |
{
|
| 135 |
+
"epoch": 4.827586206896552,
|
| 136 |
+
"grad_norm": 1.0056334733963013,
|
| 137 |
+
"learning_rate": 1.7241379310344828e-05,
|
| 138 |
+
"loss": 0.0789,
|
| 139 |
"step": 140
|
| 140 |
},
|
| 141 |
{
|
| 142 |
+
"epoch": 5.0,
|
| 143 |
+
"eval_loss": 0.04751617833971977,
|
| 144 |
+
"eval_runtime": 0.1898,
|
| 145 |
+
"eval_samples_per_second": 31.605,
|
| 146 |
+
"eval_steps_per_second": 10.535,
|
| 147 |
+
"step": 145
|
| 148 |
},
|
| 149 |
{
|
| 150 |
+
"epoch": 5.172413793103448,
|
| 151 |
+
"grad_norm": 0.9742516279220581,
|
| 152 |
+
"learning_rate": 1.704433497536946e-05,
|
| 153 |
+
"loss": 0.0927,
|
| 154 |
"step": 150
|
| 155 |
},
|
| 156 |
{
|
| 157 |
+
"epoch": 5.517241379310345,
|
| 158 |
+
"grad_norm": 1.4696099758148193,
|
| 159 |
+
"learning_rate": 1.684729064039409e-05,
|
| 160 |
+
"loss": 0.0837,
|
| 161 |
"step": 160
|
| 162 |
},
|
| 163 |
{
|
| 164 |
+
"epoch": 5.862068965517241,
|
| 165 |
+
"grad_norm": 1.0493124723434448,
|
| 166 |
+
"learning_rate": 1.665024630541872e-05,
|
| 167 |
+
"loss": 0.0689,
|
| 168 |
+
"step": 170
|
|
|
|
| 169 |
},
|
| 170 |
{
|
| 171 |
+
"epoch": 6.0,
|
| 172 |
+
"eval_loss": 0.03317258134484291,
|
| 173 |
+
"eval_runtime": 0.2045,
|
| 174 |
+
"eval_samples_per_second": 29.343,
|
| 175 |
+
"eval_steps_per_second": 9.781,
|
| 176 |
+
"step": 174
|
| 177 |
},
|
| 178 |
{
|
| 179 |
+
"epoch": 6.206896551724138,
|
| 180 |
+
"grad_norm": 0.9956067204475403,
|
| 181 |
+
"learning_rate": 1.645320197044335e-05,
|
| 182 |
+
"loss": 0.0702,
|
| 183 |
"step": 180
|
| 184 |
},
|
| 185 |
{
|
| 186 |
+
"epoch": 6.551724137931035,
|
| 187 |
+
"grad_norm": 0.4664933979511261,
|
| 188 |
+
"learning_rate": 1.625615763546798e-05,
|
| 189 |
+
"loss": 0.0675,
|
| 190 |
"step": 190
|
| 191 |
},
|
| 192 |
{
|
| 193 |
+
"epoch": 6.896551724137931,
|
| 194 |
+
"grad_norm": 1.2444266080856323,
|
| 195 |
+
"learning_rate": 1.605911330049261e-05,
|
| 196 |
+
"loss": 0.0596,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 197 |
"step": 200
|
| 198 |
},
|
| 199 |
{
|
| 200 |
+
"epoch": 7.0,
|
| 201 |
+
"eval_loss": 0.030222313478589058,
|
| 202 |
+
"eval_runtime": 0.1913,
|
| 203 |
+
"eval_samples_per_second": 31.365,
|
| 204 |
+
"eval_steps_per_second": 10.455,
|
| 205 |
+
"step": 203
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 7.241379310344827,
|
| 209 |
+
"grad_norm": 0.5432140827178955,
|
| 210 |
+
"learning_rate": 1.586206896551724e-05,
|
| 211 |
+
"loss": 0.045,
|
| 212 |
+
"step": 210
|
|
|
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 7.586206896551724,
|
| 216 |
+
"grad_norm": 0.7679450511932373,
|
| 217 |
+
"learning_rate": 1.5665024630541875e-05,
|
| 218 |
+
"loss": 0.0538,
|
| 219 |
"step": 220
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 7.931034482758621,
|
| 223 |
+
"grad_norm": 0.7759860754013062,
|
| 224 |
+
"learning_rate": 1.5467980295566506e-05,
|
| 225 |
+
"loss": 0.0624,
|
| 226 |
"step": 230
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 8.0,
|
| 230 |
+
"eval_loss": 0.02808324061334133,
|
| 231 |
+
"eval_runtime": 0.2003,
|
| 232 |
+
"eval_samples_per_second": 29.953,
|
| 233 |
+
"eval_steps_per_second": 9.984,
|
| 234 |
+
"step": 232
|
| 235 |
},
|
| 236 |
{
|
| 237 |
+
"epoch": 8.275862068965518,
|
| 238 |
+
"grad_norm": 1.7437331676483154,
|
| 239 |
+
"learning_rate": 1.5270935960591133e-05,
|
| 240 |
+
"loss": 0.0369,
|
|
|
|
| 241 |
"step": 240
|
| 242 |
},
|
| 243 |
{
|
| 244 |
+
"epoch": 8.620689655172415,
|
| 245 |
+
"grad_norm": 0.5273000597953796,
|
| 246 |
+
"learning_rate": 1.5073891625615764e-05,
|
| 247 |
+
"loss": 0.0499,
|
| 248 |
"step": 250
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 8.96551724137931,
|
| 252 |
+
"grad_norm": 0.6120426058769226,
|
| 253 |
+
"learning_rate": 1.4876847290640396e-05,
|
| 254 |
+
"loss": 0.0425,
|
| 255 |
"step": 260
|
| 256 |
},
|
| 257 |
{
|
| 258 |
+
"epoch": 9.0,
|
| 259 |
+
"eval_loss": 0.03035571426153183,
|
| 260 |
+
"eval_runtime": 0.1888,
|
| 261 |
+
"eval_samples_per_second": 31.778,
|
| 262 |
+
"eval_steps_per_second": 10.593,
|
| 263 |
+
"step": 261
|
| 264 |
},
|
| 265 |
{
|
| 266 |
+
"epoch": 9.310344827586206,
|
| 267 |
+
"grad_norm": 1.587663173675537,
|
| 268 |
+
"learning_rate": 1.4679802955665026e-05,
|
| 269 |
+
"loss": 0.0395,
|
| 270 |
"step": 270
|
| 271 |
},
|
| 272 |
{
|
| 273 |
+
"epoch": 9.655172413793103,
|
| 274 |
+
"grad_norm": 0.7260332703590393,
|
| 275 |
+
"learning_rate": 1.4482758620689657e-05,
|
| 276 |
+
"loss": 0.0493,
|
| 277 |
"step": 280
|
| 278 |
},
|
| 279 |
{
|
| 280 |
+
"epoch": 10.0,
|
| 281 |
+
"grad_norm": 0.9717508554458618,
|
| 282 |
+
"learning_rate": 1.4285714285714287e-05,
|
| 283 |
+
"loss": 0.0424,
|
| 284 |
+
"step": 290
|
|
|
|
| 285 |
},
|
| 286 |
{
|
| 287 |
+
"epoch": 10.0,
|
| 288 |
+
"eval_loss": 0.026125147938728333,
|
| 289 |
+
"eval_runtime": 0.0989,
|
| 290 |
+
"eval_samples_per_second": 60.684,
|
| 291 |
+
"eval_steps_per_second": 20.228,
|
| 292 |
"step": 290
|
| 293 |
},
|
| 294 |
{
|
| 295 |
+
"epoch": 10.344827586206897,
|
| 296 |
+
"grad_norm": 0.7487574815750122,
|
| 297 |
+
"learning_rate": 1.4088669950738918e-05,
|
| 298 |
+
"loss": 0.0412,
|
| 299 |
"step": 300
|
| 300 |
},
|
| 301 |
{
|
| 302 |
+
"epoch": 10.689655172413794,
|
| 303 |
+
"grad_norm": 1.3717066049575806,
|
| 304 |
+
"learning_rate": 1.3891625615763548e-05,
|
| 305 |
+
"loss": 0.0296,
|
| 306 |
"step": 310
|
| 307 |
},
|
| 308 |
{
|
| 309 |
+
"epoch": 11.0,
|
| 310 |
+
"eval_loss": 0.031011082231998444,
|
| 311 |
+
"eval_runtime": 0.1922,
|
| 312 |
+
"eval_samples_per_second": 31.22,
|
| 313 |
+
"eval_steps_per_second": 10.407,
|
| 314 |
+
"step": 319
|
| 315 |
},
|
| 316 |
{
|
| 317 |
+
"epoch": 11.03448275862069,
|
| 318 |
+
"grad_norm": 0.6860368847846985,
|
| 319 |
+
"learning_rate": 1.369458128078818e-05,
|
| 320 |
+
"loss": 0.0411,
|
| 321 |
"step": 320
|
| 322 |
},
|
| 323 |
{
|
| 324 |
+
"epoch": 11.379310344827585,
|
| 325 |
+
"grad_norm": 0.9999271035194397,
|
| 326 |
+
"learning_rate": 1.3497536945812807e-05,
|
| 327 |
+
"loss": 0.0358,
|
| 328 |
"step": 330
|
| 329 |
},
|
| 330 |
{
|
| 331 |
+
"epoch": 11.724137931034482,
|
| 332 |
+
"grad_norm": 1.2313721179962158,
|
| 333 |
+
"learning_rate": 1.330049261083744e-05,
|
| 334 |
+
"loss": 0.0324,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 335 |
"step": 340
|
| 336 |
},
|
| 337 |
{
|
| 338 |
+
"epoch": 12.0,
|
| 339 |
+
"eval_loss": 0.029439905658364296,
|
| 340 |
+
"eval_runtime": 0.2015,
|
| 341 |
+
"eval_samples_per_second": 29.784,
|
| 342 |
+
"eval_steps_per_second": 9.928,
|
| 343 |
+
"step": 348
|
| 344 |
},
|
| 345 |
{
|
| 346 |
+
"epoch": 12.068965517241379,
|
| 347 |
+
"grad_norm": 0.9032502174377441,
|
| 348 |
+
"learning_rate": 1.310344827586207e-05,
|
| 349 |
+
"loss": 0.041,
|
| 350 |
+
"step": 350
|
| 351 |
},
|
| 352 |
{
|
| 353 |
+
"epoch": 12.413793103448276,
|
| 354 |
+
"grad_norm": 0.612800657749176,
|
| 355 |
+
"learning_rate": 1.29064039408867e-05,
|
| 356 |
+
"loss": 0.0384,
|
|
|
|
| 357 |
"step": 360
|
| 358 |
},
|
| 359 |
{
|
| 360 |
+
"epoch": 12.758620689655173,
|
| 361 |
+
"grad_norm": 0.19512540102005005,
|
| 362 |
+
"learning_rate": 1.2709359605911331e-05,
|
| 363 |
+
"loss": 0.0349,
|
| 364 |
"step": 370
|
| 365 |
},
|
| 366 |
{
|
| 367 |
+
"epoch": 13.0,
|
| 368 |
+
"eval_loss": 0.02956104278564453,
|
| 369 |
+
"eval_runtime": 0.1909,
|
| 370 |
+
"eval_samples_per_second": 31.424,
|
| 371 |
+
"eval_steps_per_second": 10.475,
|
| 372 |
+
"step": 377
|
| 373 |
},
|
| 374 |
{
|
| 375 |
+
"epoch": 13.10344827586207,
|
| 376 |
+
"grad_norm": 0.8481155633926392,
|
| 377 |
+
"learning_rate": 1.2512315270935961e-05,
|
| 378 |
+
"loss": 0.0271,
|
| 379 |
+
"step": 380
|
|
|
|
| 380 |
},
|
| 381 |
{
|
| 382 |
+
"epoch": 13.448275862068966,
|
| 383 |
+
"grad_norm": 1.3683249950408936,
|
| 384 |
+
"learning_rate": 1.2315270935960592e-05,
|
| 385 |
+
"loss": 0.0265,
|
| 386 |
"step": 390
|
| 387 |
},
|
| 388 |
{
|
| 389 |
+
"epoch": 13.793103448275861,
|
| 390 |
+
"grad_norm": 0.6365839838981628,
|
| 391 |
+
"learning_rate": 1.2118226600985224e-05,
|
| 392 |
+
"loss": 0.0298,
|
| 393 |
"step": 400
|
| 394 |
},
|
| 395 |
{
|
| 396 |
+
"epoch": 14.0,
|
| 397 |
+
"eval_loss": 0.03038203716278076,
|
| 398 |
+
"eval_runtime": 0.1902,
|
| 399 |
+
"eval_samples_per_second": 31.538,
|
| 400 |
+
"eval_steps_per_second": 10.513,
|
| 401 |
+
"step": 406
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 14.137931034482758,
|
| 405 |
+
"grad_norm": 0.7672634124755859,
|
| 406 |
+
"learning_rate": 1.1921182266009855e-05,
|
| 407 |
+
"loss": 0.0335,
|
| 408 |
"step": 410
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"epoch": 14.482758620689655,
|
| 412 |
+
"grad_norm": 0.2541676163673401,
|
| 413 |
+
"learning_rate": 1.1724137931034483e-05,
|
| 414 |
+
"loss": 0.0286,
|
| 415 |
"step": 420
|
| 416 |
},
|
| 417 |
{
|
| 418 |
+
"epoch": 14.827586206896552,
|
| 419 |
+
"grad_norm": 0.8434980511665344,
|
| 420 |
+
"learning_rate": 1.1527093596059114e-05,
|
| 421 |
+
"loss": 0.0205,
|
| 422 |
"step": 430
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 15.0,
|
| 426 |
+
"eval_loss": 0.030394822359085083,
|
| 427 |
+
"eval_runtime": 0.19,
|
| 428 |
+
"eval_samples_per_second": 31.587,
|
| 429 |
+
"eval_steps_per_second": 10.529,
|
| 430 |
+
"step": 435
|
| 431 |
},
|
| 432 |
{
|
| 433 |
+
"epoch": 15.172413793103448,
|
| 434 |
+
"grad_norm": 0.4303562641143799,
|
| 435 |
+
"learning_rate": 1.1330049261083744e-05,
|
| 436 |
+
"loss": 0.0323,
|
| 437 |
"step": 440
|
| 438 |
},
|
| 439 |
{
|
| 440 |
+
"epoch": 15.517241379310345,
|
| 441 |
+
"grad_norm": 0.42710408568382263,
|
| 442 |
+
"learning_rate": 1.1133004926108375e-05,
|
| 443 |
+
"loss": 0.0227,
|
| 444 |
"step": 450
|
| 445 |
},
|
| 446 |
{
|
| 447 |
+
"epoch": 15.862068965517242,
|
| 448 |
+
"grad_norm": 0.6126664876937866,
|
| 449 |
+
"learning_rate": 1.0935960591133005e-05,
|
| 450 |
+
"loss": 0.0215,
|
| 451 |
+
"step": 460
|
|
|
|
| 452 |
},
|
| 453 |
{
|
| 454 |
+
"epoch": 16.0,
|
| 455 |
+
"eval_loss": 0.030341001227498055,
|
| 456 |
+
"eval_runtime": 0.1895,
|
| 457 |
+
"eval_samples_per_second": 31.669,
|
| 458 |
+
"eval_steps_per_second": 10.556,
|
| 459 |
+
"step": 464
|
| 460 |
},
|
| 461 |
{
|
| 462 |
+
"epoch": 16.20689655172414,
|
| 463 |
+
"grad_norm": 0.7011211514472961,
|
| 464 |
+
"learning_rate": 1.0738916256157637e-05,
|
| 465 |
+
"loss": 0.0337,
|
| 466 |
"step": 470
|
| 467 |
},
|
| 468 |
{
|
| 469 |
+
"epoch": 16.551724137931036,
|
| 470 |
+
"grad_norm": 0.5126092433929443,
|
| 471 |
+
"learning_rate": 1.0541871921182268e-05,
|
| 472 |
+
"loss": 0.0233,
|
| 473 |
"step": 480
|
| 474 |
},
|
| 475 |
{
|
| 476 |
+
"epoch": 16.896551724137932,
|
| 477 |
+
"grad_norm": 0.9033933281898499,
|
| 478 |
+
"learning_rate": 1.0344827586206898e-05,
|
| 479 |
+
"loss": 0.0182,
|
| 480 |
+
"step": 490
|
|
|
|
| 481 |
},
|
| 482 |
{
|
| 483 |
+
"epoch": 17.0,
|
| 484 |
+
"eval_loss": 0.027958964928984642,
|
| 485 |
+
"eval_runtime": 0.1902,
|
| 486 |
+
"eval_samples_per_second": 31.553,
|
| 487 |
+
"eval_steps_per_second": 10.518,
|
| 488 |
+
"step": 493
|
| 489 |
},
|
| 490 |
{
|
| 491 |
+
"epoch": 17.24137931034483,
|
| 492 |
+
"grad_norm": 0.4326847493648529,
|
| 493 |
+
"learning_rate": 1.0147783251231529e-05,
|
| 494 |
+
"loss": 0.0249,
|
| 495 |
"step": 500
|
| 496 |
}
|
| 497 |
],
|
| 498 |
"logging_steps": 10,
|
| 499 |
+
"max_steps": 1015,
|
| 500 |
"num_input_tokens_seen": 0,
|
| 501 |
"num_train_epochs": 35,
|
| 502 |
"save_steps": 500,
|
|
|
|
| 512 |
"attributes": {}
|
| 513 |
}
|
| 514 |
},
|
| 515 |
+
"total_flos": 1064120745984000.0,
|
| 516 |
"train_batch_size": 4,
|
| 517 |
"trial_name": null,
|
| 518 |
"trial_params": null
|
json_extraction_point_activity/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2950734544
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a3f3216f61837c04638fd20747110cf419cd0cf7777e19c57b1a79ccbd5cce8
|
| 3 |
size 2950734544
|