Commit
•
4caf23d
1
Parent(s):
7546b10
Upload folder using huggingface_hub
Browse files- config.json +29 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +869 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "xlm-roberta-large",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"XLMRobertaForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 1024,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 4096,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 514,
|
| 17 |
+
"model_type": "xlm-roberta",
|
| 18 |
+
"num_attention_heads": 16,
|
| 19 |
+
"num_hidden_layers": 24,
|
| 20 |
+
"output_past": true,
|
| 21 |
+
"pad_token_id": 1,
|
| 22 |
+
"position_embedding_type": "absolute",
|
| 23 |
+
"problem_type": "single_label_classification",
|
| 24 |
+
"torch_dtype": "float32",
|
| 25 |
+
"transformers_version": "4.41.2",
|
| 26 |
+
"type_vocab_size": 1,
|
| 27 |
+
"use_cache": true,
|
| 28 |
+
"vocab_size": 250002
|
| 29 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:daf60a4998679ac70d0ba2add1221877663daf5b7f67b9a5c703a76b4db14fdd
|
| 3 |
+
size 2239618672
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9014a78f9e7e5e45d19fa14e483cff568b36f60cb370dd7762d04e03cb73ca37
|
| 3 |
+
size 4479472721
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ca0a19ad295042159f99032b71124ba99611957add49d5a3cdcfdba689685b8
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b86637478ff97dd29675329d90ed9365d661f9d796017ebe4586552e821444e0
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,869 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8376479645847368,
|
| 3 |
+
"best_model_checkpoint": "./XLMR-large2-multi-109k-multi-outputs/checkpoint-40000",
|
| 4 |
+
"epoch": 7.893792608539648,
|
| 5 |
+
"eval_steps": 1000,
|
| 6 |
+
"global_step": 44000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.17940437746681018,
|
| 13 |
+
"grad_norm": 6.077027320861816,
|
| 14 |
+
"learning_rate": 2.242152466367713e-07,
|
| 15 |
+
"loss": 0.7043,
|
| 16 |
+
"step": 1000
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.17940437746681018,
|
| 20 |
+
"eval_accuracy": 0.5152891310929458,
|
| 21 |
+
"eval_f1": 0.4964881014781424,
|
| 22 |
+
"eval_loss": 0.6920226216316223,
|
| 23 |
+
"eval_precision": 0.522852726871274,
|
| 24 |
+
"eval_recall": 0.4726546906187625,
|
| 25 |
+
"eval_runtime": 103.5147,
|
| 26 |
+
"eval_samples_per_second": 95.726,
|
| 27 |
+
"eval_steps_per_second": 5.989,
|
| 28 |
+
"step": 1000
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 0.35880875493362036,
|
| 32 |
+
"grad_norm": 7.982357501983643,
|
| 33 |
+
"learning_rate": 4.484304932735426e-07,
|
| 34 |
+
"loss": 0.6972,
|
| 35 |
+
"step": 2000
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 0.35880875493362036,
|
| 39 |
+
"eval_accuracy": 0.5537390251286709,
|
| 40 |
+
"eval_f1": 0.514065934065934,
|
| 41 |
+
"eval_loss": 0.6867982745170593,
|
| 42 |
+
"eval_precision": 0.571882640586797,
|
| 43 |
+
"eval_recall": 0.4668662674650699,
|
| 44 |
+
"eval_runtime": 103.803,
|
| 45 |
+
"eval_samples_per_second": 95.46,
|
| 46 |
+
"eval_steps_per_second": 5.973,
|
| 47 |
+
"step": 2000
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.5382131324004306,
|
| 51 |
+
"grad_norm": 10.05902099609375,
|
| 52 |
+
"learning_rate": 6.72645739910314e-07,
|
| 53 |
+
"loss": 0.6892,
|
| 54 |
+
"step": 3000
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.5382131324004306,
|
| 58 |
+
"eval_accuracy": 0.5728125946109597,
|
| 59 |
+
"eval_f1": 0.44280637093589575,
|
| 60 |
+
"eval_loss": 0.6743206977844238,
|
| 61 |
+
"eval_precision": 0.6501739466563587,
|
| 62 |
+
"eval_recall": 0.33572854291417165,
|
| 63 |
+
"eval_runtime": 103.5733,
|
| 64 |
+
"eval_samples_per_second": 95.671,
|
| 65 |
+
"eval_steps_per_second": 5.986,
|
| 66 |
+
"step": 3000
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.7176175098672407,
|
| 70 |
+
"grad_norm": 17.559974670410156,
|
| 71 |
+
"learning_rate": 8.968609865470852e-07,
|
| 72 |
+
"loss": 0.6686,
|
| 73 |
+
"step": 4000
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.7176175098672407,
|
| 77 |
+
"eval_accuracy": 0.620345140781108,
|
| 78 |
+
"eval_f1": 0.6229705351773902,
|
| 79 |
+
"eval_loss": 0.6282544136047363,
|
| 80 |
+
"eval_precision": 0.6256038647342995,
|
| 81 |
+
"eval_recall": 0.6203592814371257,
|
| 82 |
+
"eval_runtime": 103.1262,
|
| 83 |
+
"eval_samples_per_second": 96.086,
|
| 84 |
+
"eval_steps_per_second": 6.012,
|
| 85 |
+
"step": 4000
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.897021887334051,
|
| 89 |
+
"grad_norm": 17.958587646484375,
|
| 90 |
+
"learning_rate": 9.865444034685537e-07,
|
| 91 |
+
"loss": 0.6251,
|
| 92 |
+
"step": 5000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.897021887334051,
|
| 96 |
+
"eval_accuracy": 0.6844282975073166,
|
| 97 |
+
"eval_f1": 0.671568112593215,
|
| 98 |
+
"eval_loss": 0.5713071227073669,
|
| 99 |
+
"eval_precision": 0.7087120372422966,
|
| 100 |
+
"eval_recall": 0.63812375249501,
|
| 101 |
+
"eval_runtime": 103.2452,
|
| 102 |
+
"eval_samples_per_second": 95.975,
|
| 103 |
+
"eval_steps_per_second": 6.005,
|
| 104 |
+
"step": 5000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 1.0764262648008611,
|
| 108 |
+
"grad_norm": 21.860984802246094,
|
| 109 |
+
"learning_rate": 9.616266321140236e-07,
|
| 110 |
+
"loss": 0.559,
|
| 111 |
+
"step": 6000
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 1.0764262648008611,
|
| 115 |
+
"eval_accuracy": 0.7250983953981229,
|
| 116 |
+
"eval_f1": 0.6946188340807175,
|
| 117 |
+
"eval_loss": 0.48871228098869324,
|
| 118 |
+
"eval_precision": 0.7923273657289003,
|
| 119 |
+
"eval_recall": 0.6183632734530938,
|
| 120 |
+
"eval_runtime": 102.8682,
|
| 121 |
+
"eval_samples_per_second": 96.327,
|
| 122 |
+
"eval_steps_per_second": 6.027,
|
| 123 |
+
"step": 6000
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.2558306422676713,
|
| 127 |
+
"grad_norm": 144.90872192382812,
|
| 128 |
+
"learning_rate": 9.367088607594936e-07,
|
| 129 |
+
"loss": 0.4901,
|
| 130 |
+
"step": 7000
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.2558306422676713,
|
| 134 |
+
"eval_accuracy": 0.7360984963164799,
|
| 135 |
+
"eval_f1": 0.7147998691242229,
|
| 136 |
+
"eval_loss": 0.4424116909503937,
|
| 137 |
+
"eval_precision": 0.7879297908150997,
|
| 138 |
+
"eval_recall": 0.6540918163672654,
|
| 139 |
+
"eval_runtime": 103.0989,
|
| 140 |
+
"eval_samples_per_second": 96.112,
|
| 141 |
+
"eval_steps_per_second": 6.014,
|
| 142 |
+
"step": 7000
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 1.4352350197344816,
|
| 146 |
+
"grad_norm": 23.610597610473633,
|
| 147 |
+
"learning_rate": 9.117910894049637e-07,
|
| 148 |
+
"loss": 0.4595,
|
| 149 |
+
"step": 8000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.4352350197344816,
|
| 153 |
+
"eval_accuracy": 0.7460894136643456,
|
| 154 |
+
"eval_f1": 0.7595106098260371,
|
| 155 |
+
"eval_loss": 0.4278419017791748,
|
| 156 |
+
"eval_precision": 0.7287234042553191,
|
| 157 |
+
"eval_recall": 0.7930139720558882,
|
| 158 |
+
"eval_runtime": 103.2272,
|
| 159 |
+
"eval_samples_per_second": 95.992,
|
| 160 |
+
"eval_steps_per_second": 6.006,
|
| 161 |
+
"step": 8000
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 1.6146393972012918,
|
| 165 |
+
"grad_norm": 13.438475608825684,
|
| 166 |
+
"learning_rate": 8.868733180504335e-07,
|
| 167 |
+
"loss": 0.4431,
|
| 168 |
+
"step": 9000
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.6146393972012918,
|
| 172 |
+
"eval_accuracy": 0.7517408416590978,
|
| 173 |
+
"eval_f1": 0.7542457542457542,
|
| 174 |
+
"eval_loss": 0.4201831519603729,
|
| 175 |
+
"eval_precision": 0.755,
|
| 176 |
+
"eval_recall": 0.7534930139720559,
|
| 177 |
+
"eval_runtime": 103.1397,
|
| 178 |
+
"eval_samples_per_second": 96.074,
|
| 179 |
+
"eval_steps_per_second": 6.011,
|
| 180 |
+
"step": 9000
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 1.794043774668102,
|
| 184 |
+
"grad_norm": 59.678855895996094,
|
| 185 |
+
"learning_rate": 8.619555466959035e-07,
|
| 186 |
+
"loss": 0.4246,
|
| 187 |
+
"step": 10000
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 1.794043774668102,
|
| 191 |
+
"eval_accuracy": 0.7612271672217177,
|
| 192 |
+
"eval_f1": 0.795611610228058,
|
| 193 |
+
"eval_loss": 0.4052415192127228,
|
| 194 |
+
"eval_precision": 0.7013402375875724,
|
| 195 |
+
"eval_recall": 0.9191616766467066,
|
| 196 |
+
"eval_runtime": 103.3719,
|
| 197 |
+
"eval_samples_per_second": 95.858,
|
| 198 |
+
"eval_steps_per_second": 5.998,
|
| 199 |
+
"step": 10000
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.973448152134912,
|
| 203 |
+
"grad_norm": 56.009273529052734,
|
| 204 |
+
"learning_rate": 8.370377753413735e-07,
|
| 205 |
+
"loss": 0.4168,
|
| 206 |
+
"step": 11000
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.973448152134912,
|
| 210 |
+
"eval_accuracy": 0.7611262488646685,
|
| 211 |
+
"eval_f1": 0.7569565663825855,
|
| 212 |
+
"eval_loss": 0.39738962054252625,
|
| 213 |
+
"eval_precision": 0.7794459716641996,
|
| 214 |
+
"eval_recall": 0.7357285429141717,
|
| 215 |
+
"eval_runtime": 103.343,
|
| 216 |
+
"eval_samples_per_second": 95.885,
|
| 217 |
+
"eval_steps_per_second": 5.999,
|
| 218 |
+
"step": 11000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.1528525296017222,
|
| 222 |
+
"grad_norm": 48.44904708862305,
|
| 223 |
+
"learning_rate": 8.121200039868433e-07,
|
| 224 |
+
"loss": 0.4074,
|
| 225 |
+
"step": 12000
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 2.1528525296017222,
|
| 229 |
+
"eval_accuracy": 0.7603189020082753,
|
| 230 |
+
"eval_f1": 0.7932445373030382,
|
| 231 |
+
"eval_loss": 0.42876219749450684,
|
| 232 |
+
"eval_precision": 0.7034120734908137,
|
| 233 |
+
"eval_recall": 0.9093812375249501,
|
| 234 |
+
"eval_runtime": 103.2666,
|
| 235 |
+
"eval_samples_per_second": 95.955,
|
| 236 |
+
"eval_steps_per_second": 6.004,
|
| 237 |
+
"step": 12000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.3322569070685324,
|
| 241 |
+
"grad_norm": 7.191207408905029,
|
| 242 |
+
"learning_rate": 7.872022326323134e-07,
|
| 243 |
+
"loss": 0.398,
|
| 244 |
+
"step": 13000
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 2.3322569070685324,
|
| 248 |
+
"eval_accuracy": 0.7668785952164698,
|
| 249 |
+
"eval_f1": 0.7844747154319835,
|
| 250 |
+
"eval_loss": 0.39464080333709717,
|
| 251 |
+
"eval_precision": 0.7365101611772951,
|
| 252 |
+
"eval_recall": 0.8391217564870259,
|
| 253 |
+
"eval_runtime": 103.6045,
|
| 254 |
+
"eval_samples_per_second": 95.643,
|
| 255 |
+
"eval_steps_per_second": 5.984,
|
| 256 |
+
"step": 13000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 2.5116612845353425,
|
| 260 |
+
"grad_norm": 8.779580116271973,
|
| 261 |
+
"learning_rate": 7.622844612777832e-07,
|
| 262 |
+
"loss": 0.4009,
|
| 263 |
+
"step": 14000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 2.5116612845353425,
|
| 267 |
+
"eval_accuracy": 0.7699061459279443,
|
| 268 |
+
"eval_f1": 0.7972251867662753,
|
| 269 |
+
"eval_loss": 0.38235536217689514,
|
| 270 |
+
"eval_precision": 0.7189605389797883,
|
| 271 |
+
"eval_recall": 0.8946107784431138,
|
| 272 |
+
"eval_runtime": 103.5288,
|
| 273 |
+
"eval_samples_per_second": 95.713,
|
| 274 |
+
"eval_steps_per_second": 5.989,
|
| 275 |
+
"step": 14000
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 2.6910656620021527,
|
| 279 |
+
"grad_norm": 10.75382137298584,
|
| 280 |
+
"learning_rate": 7.373666899232532e-07,
|
| 281 |
+
"loss": 0.383,
|
| 282 |
+
"step": 15000
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 2.6910656620021527,
|
| 286 |
+
"eval_accuracy": 0.7800988999899081,
|
| 287 |
+
"eval_f1": 0.7934401365058299,
|
| 288 |
+
"eval_loss": 0.4023512005805969,
|
| 289 |
+
"eval_precision": 0.7555515435999278,
|
| 290 |
+
"eval_recall": 0.8353293413173652,
|
| 291 |
+
"eval_runtime": 103.2649,
|
| 292 |
+
"eval_samples_per_second": 95.957,
|
| 293 |
+
"eval_steps_per_second": 6.004,
|
| 294 |
+
"step": 15000
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 2.8704700394689633,
|
| 298 |
+
"grad_norm": 8.595725059509277,
|
| 299 |
+
"learning_rate": 7.124489185687232e-07,
|
| 300 |
+
"loss": 0.3869,
|
| 301 |
+
"step": 16000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 2.8704700394689633,
|
| 305 |
+
"eval_accuracy": 0.7843374709859724,
|
| 306 |
+
"eval_f1": 0.7970753014908366,
|
| 307 |
+
"eval_loss": 0.3746848404407501,
|
| 308 |
+
"eval_precision": 0.7601883716717985,
|
| 309 |
+
"eval_recall": 0.8377245508982036,
|
| 310 |
+
"eval_runtime": 103.4123,
|
| 311 |
+
"eval_samples_per_second": 95.82,
|
| 312 |
+
"eval_steps_per_second": 5.995,
|
| 313 |
+
"step": 16000
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 3.0498744169357734,
|
| 317 |
+
"grad_norm": 30.062721252441406,
|
| 318 |
+
"learning_rate": 6.875311472141931e-07,
|
| 319 |
+
"loss": 0.3761,
|
| 320 |
+
"step": 17000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 3.0498744169357734,
|
| 324 |
+
"eval_accuracy": 0.7884751236249874,
|
| 325 |
+
"eval_f1": 0.7946708463949843,
|
| 326 |
+
"eval_loss": 0.39211228489875793,
|
| 327 |
+
"eval_precision": 0.7803001154290111,
|
| 328 |
+
"eval_recall": 0.8095808383233533,
|
| 329 |
+
"eval_runtime": 103.6712,
|
| 330 |
+
"eval_samples_per_second": 95.581,
|
| 331 |
+
"eval_steps_per_second": 5.98,
|
| 332 |
+
"step": 17000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 3.2292787944025836,
|
| 336 |
+
"grad_norm": 56.15926742553711,
|
| 337 |
+
"learning_rate": 6.62613375859663e-07,
|
| 338 |
+
"loss": 0.3609,
|
| 339 |
+
"step": 18000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 3.2292787944025836,
|
| 343 |
+
"eval_accuracy": 0.784741144414169,
|
| 344 |
+
"eval_f1": 0.7985645481159694,
|
| 345 |
+
"eval_loss": 0.39061158895492554,
|
| 346 |
+
"eval_precision": 0.7578419071518193,
|
| 347 |
+
"eval_recall": 0.8439121756487026,
|
| 348 |
+
"eval_runtime": 103.0596,
|
| 349 |
+
"eval_samples_per_second": 96.148,
|
| 350 |
+
"eval_steps_per_second": 6.016,
|
| 351 |
+
"step": 18000
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 3.4086831718693937,
|
| 355 |
+
"grad_norm": 7.428126811981201,
|
| 356 |
+
"learning_rate": 6.376956045051331e-07,
|
| 357 |
+
"loss": 0.3535,
|
| 358 |
+
"step": 19000
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 3.4086831718693937,
|
| 362 |
+
"eval_accuracy": 0.7953375719043294,
|
| 363 |
+
"eval_f1": 0.8157368707977467,
|
| 364 |
+
"eval_loss": 0.3811704218387604,
|
| 365 |
+
"eval_precision": 0.7486657771847899,
|
| 366 |
+
"eval_recall": 0.8960079840319362,
|
| 367 |
+
"eval_runtime": 103.3599,
|
| 368 |
+
"eval_samples_per_second": 95.869,
|
| 369 |
+
"eval_steps_per_second": 5.998,
|
| 370 |
+
"step": 19000
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 3.588087549336204,
|
| 374 |
+
"grad_norm": 248.54281616210938,
|
| 375 |
+
"learning_rate": 6.127778331506029e-07,
|
| 376 |
+
"loss": 0.3497,
|
| 377 |
+
"step": 20000
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 3.588087549336204,
|
| 381 |
+
"eval_accuracy": 0.7977596124735089,
|
| 382 |
+
"eval_f1": 0.8121836925960637,
|
| 383 |
+
"eval_loss": 0.37064263224601746,
|
| 384 |
+
"eval_precision": 0.765547703180212,
|
| 385 |
+
"eval_recall": 0.8648702594810379,
|
| 386 |
+
"eval_runtime": 103.2293,
|
| 387 |
+
"eval_samples_per_second": 95.99,
|
| 388 |
+
"eval_steps_per_second": 6.006,
|
| 389 |
+
"step": 20000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 3.767491926803014,
|
| 393 |
+
"grad_norm": 44.91804504394531,
|
| 394 |
+
"learning_rate": 5.87860061796073e-07,
|
| 395 |
+
"loss": 0.3543,
|
| 396 |
+
"step": 21000
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 3.767491926803014,
|
| 400 |
+
"eval_accuracy": 0.8025027752548188,
|
| 401 |
+
"eval_f1": 0.8141676953755579,
|
| 402 |
+
"eval_loss": 0.3442750871181488,
|
| 403 |
+
"eval_precision": 0.7764897663466763,
|
| 404 |
+
"eval_recall": 0.855688622754491,
|
| 405 |
+
"eval_runtime": 103.1871,
|
| 406 |
+
"eval_samples_per_second": 96.029,
|
| 407 |
+
"eval_steps_per_second": 6.009,
|
| 408 |
+
"step": 21000
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 3.946896304269824,
|
| 412 |
+
"grad_norm": 84.26334381103516,
|
| 413 |
+
"learning_rate": 5.629422904415428e-07,
|
| 414 |
+
"loss": 0.3425,
|
| 415 |
+
"step": 22000
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 3.946896304269824,
|
| 419 |
+
"eval_accuracy": 0.8035119588253103,
|
| 420 |
+
"eval_f1": 0.8079692277344905,
|
| 421 |
+
"eval_loss": 0.3556448519229889,
|
| 422 |
+
"eval_precision": 0.7985962175862741,
|
| 423 |
+
"eval_recall": 0.817564870259481,
|
| 424 |
+
"eval_runtime": 102.9714,
|
| 425 |
+
"eval_samples_per_second": 96.231,
|
| 426 |
+
"eval_steps_per_second": 6.021,
|
| 427 |
+
"step": 22000
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 4.126300681736635,
|
| 431 |
+
"grad_norm": 23.619245529174805,
|
| 432 |
+
"learning_rate": 5.380245190870128e-07,
|
| 433 |
+
"loss": 0.335,
|
| 434 |
+
"step": 23000
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 4.126300681736635,
|
| 438 |
+
"eval_accuracy": 0.8024018568977697,
|
| 439 |
+
"eval_f1": 0.8052903739061257,
|
| 440 |
+
"eval_loss": 0.3544567823410034,
|
| 441 |
+
"eval_precision": 0.8024177566389219,
|
| 442 |
+
"eval_recall": 0.808183632734531,
|
| 443 |
+
"eval_runtime": 102.8733,
|
| 444 |
+
"eval_samples_per_second": 96.322,
|
| 445 |
+
"eval_steps_per_second": 6.027,
|
| 446 |
+
"step": 23000
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 4.3057050592034445,
|
| 450 |
+
"grad_norm": 16.840389251708984,
|
| 451 |
+
"learning_rate": 5.131067477324828e-07,
|
| 452 |
+
"loss": 0.3222,
|
| 453 |
+
"step": 24000
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 4.3057050592034445,
|
| 457 |
+
"eval_accuracy": 0.8070441013220305,
|
| 458 |
+
"eval_f1": 0.8103550882761357,
|
| 459 |
+
"eval_loss": 0.349142849445343,
|
| 460 |
+
"eval_precision": 0.8054022082018928,
|
| 461 |
+
"eval_recall": 0.8153692614770459,
|
| 462 |
+
"eval_runtime": 103.0742,
|
| 463 |
+
"eval_samples_per_second": 96.135,
|
| 464 |
+
"eval_steps_per_second": 6.015,
|
| 465 |
+
"step": 24000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 4.485109436670255,
|
| 469 |
+
"grad_norm": 49.83803939819336,
|
| 470 |
+
"learning_rate": 4.881889763779527e-07,
|
| 471 |
+
"loss": 0.3157,
|
| 472 |
+
"step": 25000
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 4.485109436670255,
|
| 476 |
+
"eval_accuracy": 0.8095670602482592,
|
| 477 |
+
"eval_f1": 0.8254232583957813,
|
| 478 |
+
"eval_loss": 0.357431560754776,
|
| 479 |
+
"eval_precision": 0.7692705638903259,
|
| 480 |
+
"eval_recall": 0.8904191616766467,
|
| 481 |
+
"eval_runtime": 103.3608,
|
| 482 |
+
"eval_samples_per_second": 95.868,
|
| 483 |
+
"eval_steps_per_second": 5.998,
|
| 484 |
+
"step": 25000
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 4.664513814137065,
|
| 488 |
+
"grad_norm": 134.8468475341797,
|
| 489 |
+
"learning_rate": 4.632712050234227e-07,
|
| 490 |
+
"loss": 0.3207,
|
| 491 |
+
"step": 26000
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 4.664513814137065,
|
| 495 |
+
"eval_accuracy": 0.8153194066000605,
|
| 496 |
+
"eval_f1": 0.8328156404165905,
|
| 497 |
+
"eval_loss": 0.34428831934928894,
|
| 498 |
+
"eval_precision": 0.7678571428571429,
|
| 499 |
+
"eval_recall": 0.9097804391217564,
|
| 500 |
+
"eval_runtime": 103.0601,
|
| 501 |
+
"eval_samples_per_second": 96.148,
|
| 502 |
+
"eval_steps_per_second": 6.016,
|
| 503 |
+
"step": 26000
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 4.843918191603875,
|
| 507 |
+
"grad_norm": 12.487037658691406,
|
| 508 |
+
"learning_rate": 4.3835343366889267e-07,
|
| 509 |
+
"loss": 0.3217,
|
| 510 |
+
"step": 27000
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 4.843918191603875,
|
| 514 |
+
"eval_accuracy": 0.8124936926026844,
|
| 515 |
+
"eval_f1": 0.8151611619578193,
|
| 516 |
+
"eval_loss": 0.3367626368999481,
|
| 517 |
+
"eval_precision": 0.8125743752479175,
|
| 518 |
+
"eval_recall": 0.8177644710578842,
|
| 519 |
+
"eval_runtime": 103.075,
|
| 520 |
+
"eval_samples_per_second": 96.134,
|
| 521 |
+
"eval_steps_per_second": 6.015,
|
| 522 |
+
"step": 27000
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 5.023322569070685,
|
| 526 |
+
"grad_norm": 10.074256896972656,
|
| 527 |
+
"learning_rate": 4.1343566231436264e-07,
|
| 528 |
+
"loss": 0.3184,
|
| 529 |
+
"step": 28000
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 5.023322569070685,
|
| 533 |
+
"eval_accuracy": 0.8171359370269452,
|
| 534 |
+
"eval_f1": 0.8204518430439952,
|
| 535 |
+
"eval_loss": 0.3432736396789551,
|
| 536 |
+
"eval_precision": 0.8146399055489965,
|
| 537 |
+
"eval_recall": 0.8263473053892215,
|
| 538 |
+
"eval_runtime": 103.2718,
|
| 539 |
+
"eval_samples_per_second": 95.951,
|
| 540 |
+
"eval_steps_per_second": 6.004,
|
| 541 |
+
"step": 28000
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 5.202726946537496,
|
| 545 |
+
"grad_norm": 30.08102035522461,
|
| 546 |
+
"learning_rate": 3.8851789095983255e-07,
|
| 547 |
+
"loss": 0.2981,
|
| 548 |
+
"step": 29000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 5.202726946537496,
|
| 552 |
+
"eval_accuracy": 0.8162276718135029,
|
| 553 |
+
"eval_f1": 0.8155202107182656,
|
| 554 |
+
"eval_loss": 0.34637027978897095,
|
| 555 |
+
"eval_precision": 0.8280189261468833,
|
| 556 |
+
"eval_recall": 0.8033932135728543,
|
| 557 |
+
"eval_runtime": 102.9045,
|
| 558 |
+
"eval_samples_per_second": 96.293,
|
| 559 |
+
"eval_steps_per_second": 6.025,
|
| 560 |
+
"step": 29000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 5.382131324004305,
|
| 564 |
+
"grad_norm": 12.194862365722656,
|
| 565 |
+
"learning_rate": 3.6360011960530246e-07,
|
| 566 |
+
"loss": 0.301,
|
| 567 |
+
"step": 30000
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 5.382131324004305,
|
| 571 |
+
"eval_accuracy": 0.817640528812191,
|
| 572 |
+
"eval_f1": 0.8211776348342404,
|
| 573 |
+
"eval_loss": 0.342290461063385,
|
| 574 |
+
"eval_precision": 0.8143277723258097,
|
| 575 |
+
"eval_recall": 0.8281437125748503,
|
| 576 |
+
"eval_runtime": 103.5562,
|
| 577 |
+
"eval_samples_per_second": 95.687,
|
| 578 |
+
"eval_steps_per_second": 5.987,
|
| 579 |
+
"step": 30000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 5.561535701471116,
|
| 583 |
+
"grad_norm": 154.16159057617188,
|
| 584 |
+
"learning_rate": 3.386823482507724e-07,
|
| 585 |
+
"loss": 0.2979,
|
| 586 |
+
"step": 31000
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 5.561535701471116,
|
| 590 |
+
"eval_accuracy": 0.8201634877384196,
|
| 591 |
+
"eval_f1": 0.8214070956103428,
|
| 592 |
+
"eval_loss": 0.32883062958717346,
|
| 593 |
+
"eval_precision": 0.8248792270531401,
|
| 594 |
+
"eval_recall": 0.8179640718562874,
|
| 595 |
+
"eval_runtime": 107.2514,
|
| 596 |
+
"eval_samples_per_second": 92.39,
|
| 597 |
+
"eval_steps_per_second": 5.781,
|
| 598 |
+
"step": 31000
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 5.740940078937927,
|
| 602 |
+
"grad_norm": 20.60382080078125,
|
| 603 |
+
"learning_rate": 3.137645768962424e-07,
|
| 604 |
+
"loss": 0.2941,
|
| 605 |
+
"step": 32000
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 5.740940078937927,
|
| 609 |
+
"eval_accuracy": 0.8215763447371077,
|
| 610 |
+
"eval_f1": 0.8254689042448173,
|
| 611 |
+
"eval_loss": 0.341677188873291,
|
| 612 |
+
"eval_precision": 0.8166015625,
|
| 613 |
+
"eval_recall": 0.8345309381237525,
|
| 614 |
+
"eval_runtime": 103.1122,
|
| 615 |
+
"eval_samples_per_second": 96.099,
|
| 616 |
+
"eval_steps_per_second": 6.013,
|
| 617 |
+
"step": 32000
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 5.920344456404736,
|
| 621 |
+
"grad_norm": 27.749670028686523,
|
| 622 |
+
"learning_rate": 2.888468055417123e-07,
|
| 623 |
+
"loss": 0.3015,
|
| 624 |
+
"step": 33000
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 5.920344456404736,
|
| 628 |
+
"eval_accuracy": 0.8243011403774346,
|
| 629 |
+
"eval_f1": 0.8335404914427765,
|
| 630 |
+
"eval_loss": 0.33678942918777466,
|
| 631 |
+
"eval_precision": 0.799963296017618,
|
| 632 |
+
"eval_recall": 0.870059880239521,
|
| 633 |
+
"eval_runtime": 103.0115,
|
| 634 |
+
"eval_samples_per_second": 96.193,
|
| 635 |
+
"eval_steps_per_second": 6.019,
|
| 636 |
+
"step": 33000
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 6.099748833871547,
|
| 640 |
+
"grad_norm": 63.67295455932617,
|
| 641 |
+
"learning_rate": 2.6392903418718226e-07,
|
| 642 |
+
"loss": 0.2953,
|
| 643 |
+
"step": 34000
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 6.099748833871547,
|
| 647 |
+
"eval_accuracy": 0.8256130790190735,
|
| 648 |
+
"eval_f1": 0.8240684178375076,
|
| 649 |
+
"eval_loss": 0.33581623435020447,
|
| 650 |
+
"eval_precision": 0.8410224438902744,
|
| 651 |
+
"eval_recall": 0.8077844311377246,
|
| 652 |
+
"eval_runtime": 103.1426,
|
| 653 |
+
"eval_samples_per_second": 96.071,
|
| 654 |
+
"eval_steps_per_second": 6.011,
|
| 655 |
+
"step": 34000
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 6.279153211338357,
|
| 659 |
+
"grad_norm": 26.843647003173828,
|
| 660 |
+
"learning_rate": 2.390112628326522e-07,
|
| 661 |
+
"loss": 0.2852,
|
| 662 |
+
"step": 35000
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 6.279153211338357,
|
| 666 |
+
"eval_accuracy": 0.8249066505197296,
|
| 667 |
+
"eval_f1": 0.8327065856715842,
|
| 668 |
+
"eval_loss": 0.34431934356689453,
|
| 669 |
+
"eval_precision": 0.8054467450102593,
|
| 670 |
+
"eval_recall": 0.86187624750499,
|
| 671 |
+
"eval_runtime": 103.3497,
|
| 672 |
+
"eval_samples_per_second": 95.878,
|
| 673 |
+
"eval_steps_per_second": 5.999,
|
| 674 |
+
"step": 35000
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 6.458557588805167,
|
| 678 |
+
"grad_norm": 183.19422912597656,
|
| 679 |
+
"learning_rate": 2.140934914781222e-07,
|
| 680 |
+
"loss": 0.2917,
|
| 681 |
+
"step": 36000
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 6.458557588805167,
|
| 685 |
+
"eval_accuracy": 0.824502977091533,
|
| 686 |
+
"eval_f1": 0.8318669631634922,
|
| 687 |
+
"eval_loss": 0.34868115186691284,
|
| 688 |
+
"eval_precision": 0.8066754172135758,
|
| 689 |
+
"eval_recall": 0.858682634730539,
|
| 690 |
+
"eval_runtime": 103.3302,
|
| 691 |
+
"eval_samples_per_second": 95.897,
|
| 692 |
+
"eval_steps_per_second": 6.0,
|
| 693 |
+
"step": 36000
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 6.637961966271977,
|
| 697 |
+
"grad_norm": 10.319212913513184,
|
| 698 |
+
"learning_rate": 1.8917572012359216e-07,
|
| 699 |
+
"loss": 0.2844,
|
| 700 |
+
"step": 37000
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 6.637961966271977,
|
| 704 |
+
"eval_accuracy": 0.8261176708043193,
|
| 705 |
+
"eval_f1": 0.8294565970503811,
|
| 706 |
+
"eval_loss": 0.32437703013420105,
|
| 707 |
+
"eval_precision": 0.8226978205379933,
|
| 708 |
+
"eval_recall": 0.8363273453093812,
|
| 709 |
+
"eval_runtime": 104.1332,
|
| 710 |
+
"eval_samples_per_second": 95.157,
|
| 711 |
+
"eval_steps_per_second": 5.954,
|
| 712 |
+
"step": 37000
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 6.8173663437387875,
|
| 716 |
+
"grad_norm": 58.22975540161133,
|
| 717 |
+
"learning_rate": 1.642579487690621e-07,
|
| 718 |
+
"loss": 0.2837,
|
| 719 |
+
"step": 38000
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 6.8173663437387875,
|
| 723 |
+
"eval_accuracy": 0.8285397113734988,
|
| 724 |
+
"eval_f1": 0.8333823673629499,
|
| 725 |
+
"eval_loss": 0.3295113742351532,
|
| 726 |
+
"eval_precision": 0.8191632928475033,
|
| 727 |
+
"eval_recall": 0.8481037924151696,
|
| 728 |
+
"eval_runtime": 103.1218,
|
| 729 |
+
"eval_samples_per_second": 96.09,
|
| 730 |
+
"eval_steps_per_second": 6.012,
|
| 731 |
+
"step": 38000
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 6.996770721205597,
|
| 735 |
+
"grad_norm": 8.232932090759277,
|
| 736 |
+
"learning_rate": 1.3934017741453206e-07,
|
| 737 |
+
"loss": 0.283,
|
| 738 |
+
"step": 39000
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 6.996770721205597,
|
| 742 |
+
"eval_accuracy": 0.8263195075184177,
|
| 743 |
+
"eval_f1": 0.8296882731321128,
|
| 744 |
+
"eval_loss": 0.3371128439903259,
|
| 745 |
+
"eval_precision": 0.8227674190382728,
|
| 746 |
+
"eval_recall": 0.8367265469061876,
|
| 747 |
+
"eval_runtime": 103.2754,
|
| 748 |
+
"eval_samples_per_second": 95.947,
|
| 749 |
+
"eval_steps_per_second": 6.003,
|
| 750 |
+
"step": 39000
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 7.176175098672408,
|
| 754 |
+
"grad_norm": 18.62181282043457,
|
| 755 |
+
"learning_rate": 1.14422406060002e-07,
|
| 756 |
+
"loss": 0.2711,
|
| 757 |
+
"step": 40000
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 7.176175098672408,
|
| 761 |
+
"eval_accuracy": 0.8297507316580887,
|
| 762 |
+
"eval_f1": 0.8376479645847368,
|
| 763 |
+
"eval_loss": 0.32895320653915405,
|
| 764 |
+
"eval_precision": 0.8087716037911169,
|
| 765 |
+
"eval_recall": 0.8686626746506986,
|
| 766 |
+
"eval_runtime": 103.3849,
|
| 767 |
+
"eval_samples_per_second": 95.846,
|
| 768 |
+
"eval_steps_per_second": 5.997,
|
| 769 |
+
"step": 40000
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 7.3555794761392175,
|
| 773 |
+
"grad_norm": 15.900300025939941,
|
| 774 |
+
"learning_rate": 8.950463470547195e-08,
|
| 775 |
+
"loss": 0.273,
|
| 776 |
+
"step": 41000
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 7.3555794761392175,
|
| 780 |
+
"eval_accuracy": 0.8292461398728429,
|
| 781 |
+
"eval_f1": 0.8348946135831382,
|
| 782 |
+
"eval_loss": 0.34222128987312317,
|
| 783 |
+
"eval_precision": 0.8167239404352806,
|
| 784 |
+
"eval_recall": 0.8538922155688623,
|
| 785 |
+
"eval_runtime": 103.3124,
|
| 786 |
+
"eval_samples_per_second": 95.913,
|
| 787 |
+
"eval_steps_per_second": 6.001,
|
| 788 |
+
"step": 41000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 7.534983853606028,
|
| 792 |
+
"grad_norm": 54.62172317504883,
|
| 793 |
+
"learning_rate": 6.45868633509419e-08,
|
| 794 |
+
"loss": 0.2795,
|
| 795 |
+
"step": 42000
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 7.534983853606028,
|
| 799 |
+
"eval_accuracy": 0.8275305278030074,
|
| 800 |
+
"eval_f1": 0.8291512546236129,
|
| 801 |
+
"eval_loss": 0.33169299364089966,
|
| 802 |
+
"eval_precision": 0.8305627879030643,
|
| 803 |
+
"eval_recall": 0.8277445109780439,
|
| 804 |
+
"eval_runtime": 103.4355,
|
| 805 |
+
"eval_samples_per_second": 95.799,
|
| 806 |
+
"eval_steps_per_second": 5.994,
|
| 807 |
+
"step": 42000
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 7.714388231072839,
|
| 811 |
+
"grad_norm": 47.589847564697266,
|
| 812 |
+
"learning_rate": 3.9669091996411835e-08,
|
| 813 |
+
"loss": 0.2739,
|
| 814 |
+
"step": 43000
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 7.714388231072839,
|
| 818 |
+
"eval_accuracy": 0.8305580785144818,
|
| 819 |
+
"eval_f1": 0.8372904351196822,
|
| 820 |
+
"eval_loss": 0.336332768201828,
|
| 821 |
+
"eval_precision": 0.8137125635712941,
|
| 822 |
+
"eval_recall": 0.8622754491017964,
|
| 823 |
+
"eval_runtime": 103.0445,
|
| 824 |
+
"eval_samples_per_second": 96.162,
|
| 825 |
+
"eval_steps_per_second": 6.017,
|
| 826 |
+
"step": 43000
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 7.893792608539648,
|
| 830 |
+
"grad_norm": 32.284854888916016,
|
| 831 |
+
"learning_rate": 1.475132064188179e-08,
|
| 832 |
+
"loss": 0.2771,
|
| 833 |
+
"step": 44000
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 7.893792608539648,
|
| 837 |
+
"eval_accuracy": 0.8294479765869411,
|
| 838 |
+
"eval_f1": 0.8322747121873759,
|
| 839 |
+
"eval_loss": 0.33281558752059937,
|
| 840 |
+
"eval_precision": 0.8276746940386893,
|
| 841 |
+
"eval_recall": 0.8369261477045908,
|
| 842 |
+
"eval_runtime": 103.2292,
|
| 843 |
+
"eval_samples_per_second": 95.99,
|
| 844 |
+
"eval_steps_per_second": 6.006,
|
| 845 |
+
"step": 44000
|
| 846 |
+
}
|
| 847 |
+
],
|
| 848 |
+
"logging_steps": 1000,
|
| 849 |
+
"max_steps": 44592,
|
| 850 |
+
"num_input_tokens_seen": 0,
|
| 851 |
+
"num_train_epochs": 8,
|
| 852 |
+
"save_steps": 1000,
|
| 853 |
+
"stateful_callbacks": {
|
| 854 |
+
"TrainerControl": {
|
| 855 |
+
"args": {
|
| 856 |
+
"should_epoch_stop": false,
|
| 857 |
+
"should_evaluate": false,
|
| 858 |
+
"should_log": false,
|
| 859 |
+
"should_save": true,
|
| 860 |
+
"should_training_stop": false
|
| 861 |
+
},
|
| 862 |
+
"attributes": {}
|
| 863 |
+
}
|
| 864 |
+
},
|
| 865 |
+
"total_flos": 4.949009928618441e+17,
|
| 866 |
+
"train_batch_size": 16,
|
| 867 |
+
"trial_name": null,
|
| 868 |
+
"trial_params": null
|
| 869 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:74bc508abee099291ad8894c55837db6477c06aa2700860858f3c6e09ed6af95
|
| 3 |
+
size 5112
|