Upload folder using huggingface_hub
Browse files- checkpoints/checkpoint-4966/config.json +26 -0
- checkpoints/checkpoint-4966/model.safetensors +3 -0
- checkpoints/checkpoint-4966/optimizer.pt +3 -0
- checkpoints/checkpoint-4966/rng_state.pth +3 -0
- checkpoints/checkpoint-4966/scheduler.pt +3 -0
- checkpoints/checkpoint-4966/trainer_state.json +826 -0
- checkpoints/checkpoint-4966/training_args.bin +3 -0
checkpoints/checkpoint-4966/config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForMaskedLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"layer_norm_eps": 1e-05,
|
| 15 |
+
"max_position_embeddings": 514,
|
| 16 |
+
"model_type": "roberta",
|
| 17 |
+
"num_attention_heads": 12,
|
| 18 |
+
"num_hidden_layers": 12,
|
| 19 |
+
"pad_token_id": 1,
|
| 20 |
+
"position_embedding_type": "absolute",
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.47.0",
|
| 23 |
+
"type_vocab_size": 1,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 52000
|
| 26 |
+
}
|
checkpoints/checkpoint-4966/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e157fc53be3295298fff26745a4e98da91fc87e945d30c00328e8318ae9b6069
|
| 3 |
+
size 504150808
|
checkpoints/checkpoint-4966/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:31cbcb460a0753dc627276d637d05ad23ec5dba485c44aac1e71f8d3a36cf3df
|
| 3 |
+
size 1008422138
|
checkpoints/checkpoint-4966/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2c9bd5dd139ab5218d2a176674ad5efa936f8c963f8d79e0e01fae21d512c834
|
| 3 |
+
size 14244
|
checkpoints/checkpoint-4966/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c2ccbc5ec934dd650192ed9e7bb73a559229bfde8832431cf459f508f961cff1
|
| 3 |
+
size 1064
|
checkpoints/checkpoint-4966/trainer_state.json
ADDED
|
@@ -0,0 +1,826 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 9.998697351280939,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 4966,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.08684324793747286,
|
| 13 |
+
"grad_norm": 48.32429885864258,
|
| 14 |
+
"learning_rate": 2.5e-06,
|
| 15 |
+
"loss": 10.7455,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.17368649587494572,
|
| 20 |
+
"grad_norm": 36.34796142578125,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 10.0183,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.26052974381241856,
|
| 27 |
+
"grad_norm": 33.813438415527344,
|
| 28 |
+
"learning_rate": 7.5e-06,
|
| 29 |
+
"loss": 9.6101,
|
| 30 |
+
"step": 150
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.34737299174989145,
|
| 34 |
+
"grad_norm": 32.9961051940918,
|
| 35 |
+
"learning_rate": 1e-05,
|
| 36 |
+
"loss": 9.1957,
|
| 37 |
+
"step": 200
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.4342162396873643,
|
| 41 |
+
"grad_norm": 32.04951858520508,
|
| 42 |
+
"learning_rate": 1.25e-05,
|
| 43 |
+
"loss": 8.7435,
|
| 44 |
+
"step": 250
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.5210594876248371,
|
| 48 |
+
"grad_norm": 27.619997024536133,
|
| 49 |
+
"learning_rate": 1.5e-05,
|
| 50 |
+
"loss": 8.2703,
|
| 51 |
+
"step": 300
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.60790273556231,
|
| 55 |
+
"grad_norm": 34.229496002197266,
|
| 56 |
+
"learning_rate": 1.7500000000000002e-05,
|
| 57 |
+
"loss": 7.8107,
|
| 58 |
+
"step": 350
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.6947459834997829,
|
| 62 |
+
"grad_norm": 24.013845443725586,
|
| 63 |
+
"learning_rate": 2e-05,
|
| 64 |
+
"loss": 7.3793,
|
| 65 |
+
"step": 400
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.7815892314372558,
|
| 69 |
+
"grad_norm": 22.049392700195312,
|
| 70 |
+
"learning_rate": 2.2499999999999998e-05,
|
| 71 |
+
"loss": 7.064,
|
| 72 |
+
"step": 450
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.7989578810247503,
|
| 76 |
+
"eval_accuracy": 0.0,
|
| 77 |
+
"eval_loss": 6.8795037269592285,
|
| 78 |
+
"eval_normalizer": 136811.0,
|
| 79 |
+
"eval_runtime": 153.0785,
|
| 80 |
+
"eval_samples_per_second": 385.096,
|
| 81 |
+
"eval_steps_per_second": 0.758,
|
| 82 |
+
"step": 460
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 1.0694745983499783,
|
| 86 |
+
"grad_norm": 17.628145217895508,
|
| 87 |
+
"learning_rate": 2.5e-05,
|
| 88 |
+
"loss": 6.9743,
|
| 89 |
+
"step": 500
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.1563178462874513,
|
| 93 |
+
"grad_norm": 24.724470138549805,
|
| 94 |
+
"learning_rate": 2.75e-05,
|
| 95 |
+
"loss": 6.7219,
|
| 96 |
+
"step": 550
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.243161094224924,
|
| 100 |
+
"grad_norm": 26.161134719848633,
|
| 101 |
+
"learning_rate": 3e-05,
|
| 102 |
+
"loss": 6.6322,
|
| 103 |
+
"step": 600
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 1.3300043421623968,
|
| 107 |
+
"grad_norm": 23.512306213378906,
|
| 108 |
+
"learning_rate": 3.2500000000000004e-05,
|
| 109 |
+
"loss": 6.5439,
|
| 110 |
+
"step": 650
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 1.4168475900998696,
|
| 114 |
+
"grad_norm": 26.923786163330078,
|
| 115 |
+
"learning_rate": 3.5000000000000004e-05,
|
| 116 |
+
"loss": 6.4634,
|
| 117 |
+
"step": 700
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 1.5036908380373426,
|
| 121 |
+
"grad_norm": 23.046533584594727,
|
| 122 |
+
"learning_rate": 3.75e-05,
|
| 123 |
+
"loss": 6.3879,
|
| 124 |
+
"step": 750
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 1.5905340859748156,
|
| 128 |
+
"grad_norm": 22.831220626831055,
|
| 129 |
+
"learning_rate": 4e-05,
|
| 130 |
+
"loss": 6.3118,
|
| 131 |
+
"step": 800
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 1.6356925749023015,
|
| 135 |
+
"eval_accuracy": 0.0,
|
| 136 |
+
"eval_loss": 5.832508087158203,
|
| 137 |
+
"eval_normalizer": 136797.0,
|
| 138 |
+
"eval_runtime": 147.1989,
|
| 139 |
+
"eval_samples_per_second": 400.478,
|
| 140 |
+
"eval_steps_per_second": 0.788,
|
| 141 |
+
"step": 826
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 2.041684759009987,
|
| 145 |
+
"grad_norm": 22.007122039794922,
|
| 146 |
+
"learning_rate": 4.25e-05,
|
| 147 |
+
"loss": 6.2933,
|
| 148 |
+
"step": 850
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 2.1285280069474597,
|
| 152 |
+
"grad_norm": 16.270891189575195,
|
| 153 |
+
"learning_rate": 4.4999999999999996e-05,
|
| 154 |
+
"loss": 6.1964,
|
| 155 |
+
"step": 900
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 2.215371254884933,
|
| 159 |
+
"grad_norm": 22.500335693359375,
|
| 160 |
+
"learning_rate": 4.75e-05,
|
| 161 |
+
"loss": 6.1024,
|
| 162 |
+
"step": 950
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 2.3022145028224057,
|
| 166 |
+
"grad_norm": 24.74261474609375,
|
| 167 |
+
"learning_rate": 5e-05,
|
| 168 |
+
"loss": 6.0502,
|
| 169 |
+
"step": 1000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 2.3890577507598785,
|
| 173 |
+
"grad_norm": 31.56224822998047,
|
| 174 |
+
"learning_rate": 5.25e-05,
|
| 175 |
+
"loss": 5.9904,
|
| 176 |
+
"step": 1050
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 2.4759009986973513,
|
| 180 |
+
"grad_norm": 18.931520462036133,
|
| 181 |
+
"learning_rate": 5.5e-05,
|
| 182 |
+
"loss": 5.9306,
|
| 183 |
+
"step": 1100
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 2.528006947459835,
|
| 187 |
+
"eval_accuracy": 0.0,
|
| 188 |
+
"eval_loss": 5.32896089553833,
|
| 189 |
+
"eval_normalizer": 135800.0,
|
| 190 |
+
"eval_runtime": 128.1308,
|
| 191 |
+
"eval_samples_per_second": 460.077,
|
| 192 |
+
"eval_steps_per_second": 0.905,
|
| 193 |
+
"step": 1130
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 3.034737299174989,
|
| 197 |
+
"grad_norm": 25.570791244506836,
|
| 198 |
+
"learning_rate": 5.75e-05,
|
| 199 |
+
"loss": 5.7977,
|
| 200 |
+
"step": 1150
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 3.121580547112462,
|
| 204 |
+
"grad_norm": 29.18811798095703,
|
| 205 |
+
"learning_rate": 6e-05,
|
| 206 |
+
"loss": 5.5845,
|
| 207 |
+
"step": 1200
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 3.208423795049935,
|
| 211 |
+
"grad_norm": 23.236276626586914,
|
| 212 |
+
"learning_rate": 6.25e-05,
|
| 213 |
+
"loss": 5.4957,
|
| 214 |
+
"step": 1250
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 3.295267042987408,
|
| 218 |
+
"grad_norm": 23.984031677246094,
|
| 219 |
+
"learning_rate": 6.500000000000001e-05,
|
| 220 |
+
"loss": 5.4124,
|
| 221 |
+
"step": 1300
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 3.3821102909248806,
|
| 225 |
+
"grad_norm": 28.23763656616211,
|
| 226 |
+
"learning_rate": 6.75e-05,
|
| 227 |
+
"loss": 5.323,
|
| 228 |
+
"step": 1350
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 3.4689535388623534,
|
| 232 |
+
"grad_norm": 30.71808624267578,
|
| 233 |
+
"learning_rate": 7.000000000000001e-05,
|
| 234 |
+
"loss": 5.218,
|
| 235 |
+
"step": 1400
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 3.555796786799826,
|
| 239 |
+
"grad_norm": 32.23666763305664,
|
| 240 |
+
"learning_rate": 7.25e-05,
|
| 241 |
+
"loss": 5.1124,
|
| 242 |
+
"step": 1450
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 3.6426400347372994,
|
| 246 |
+
"grad_norm": 28.51193618774414,
|
| 247 |
+
"learning_rate": 7.5e-05,
|
| 248 |
+
"loss": 5.0115,
|
| 249 |
+
"step": 1500
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 3.6704298740772905,
|
| 253 |
+
"eval_accuracy": 0.0,
|
| 254 |
+
"eval_loss": 4.709470748901367,
|
| 255 |
+
"eval_normalizer": 136953.0,
|
| 256 |
+
"eval_runtime": 107.2678,
|
| 257 |
+
"eval_samples_per_second": 549.559,
|
| 258 |
+
"eval_steps_per_second": 1.081,
|
| 259 |
+
"step": 1516
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 4.059053408597482,
|
| 263 |
+
"grad_norm": 26.30297088623047,
|
| 264 |
+
"learning_rate": 7.75e-05,
|
| 265 |
+
"loss": 4.5805,
|
| 266 |
+
"step": 1550
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 4.145896656534954,
|
| 270 |
+
"grad_norm": 26.367795944213867,
|
| 271 |
+
"learning_rate": 8e-05,
|
| 272 |
+
"loss": 4.2572,
|
| 273 |
+
"step": 1600
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 4.232739904472427,
|
| 277 |
+
"grad_norm": 27.613256454467773,
|
| 278 |
+
"learning_rate": 8.25e-05,
|
| 279 |
+
"loss": 4.1758,
|
| 280 |
+
"step": 1650
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 4.3195831524099,
|
| 284 |
+
"grad_norm": 26.149005889892578,
|
| 285 |
+
"learning_rate": 8.5e-05,
|
| 286 |
+
"loss": 4.1026,
|
| 287 |
+
"step": 1700
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 4.406426400347373,
|
| 291 |
+
"grad_norm": 25.32581329345703,
|
| 292 |
+
"learning_rate": 8.75e-05,
|
| 293 |
+
"loss": 4.0214,
|
| 294 |
+
"step": 1750
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 4.493269648284846,
|
| 298 |
+
"grad_norm": 25.05300521850586,
|
| 299 |
+
"learning_rate": 8.999999999999999e-05,
|
| 300 |
+
"loss": 3.9594,
|
| 301 |
+
"step": 1800
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 4.580112896222319,
|
| 305 |
+
"grad_norm": 26.33847427368164,
|
| 306 |
+
"learning_rate": 9.25e-05,
|
| 307 |
+
"loss": 3.9085,
|
| 308 |
+
"step": 1850
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 4.6669561441597915,
|
| 312 |
+
"grad_norm": 23.64914894104004,
|
| 313 |
+
"learning_rate": 9.5e-05,
|
| 314 |
+
"loss": 3.8477,
|
| 315 |
+
"step": 1900
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 4.753799392097265,
|
| 319 |
+
"grad_norm": 25.411096572875977,
|
| 320 |
+
"learning_rate": 9.750000000000001e-05,
|
| 321 |
+
"loss": 3.8064,
|
| 322 |
+
"step": 1950
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 4.840642640034737,
|
| 326 |
+
"grad_norm": 25.952960968017578,
|
| 327 |
+
"learning_rate": 0.0001,
|
| 328 |
+
"loss": 3.7317,
|
| 329 |
+
"step": 2000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 4.92748588797221,
|
| 333 |
+
"grad_norm": 27.8713321685791,
|
| 334 |
+
"learning_rate": 0.0001025,
|
| 335 |
+
"loss": 3.7005,
|
| 336 |
+
"step": 2050
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 4.998697351280938,
|
| 340 |
+
"eval_accuracy": 0.0,
|
| 341 |
+
"eval_loss": 4.281714916229248,
|
| 342 |
+
"eval_normalizer": 136898.0,
|
| 343 |
+
"eval_runtime": 104.4121,
|
| 344 |
+
"eval_samples_per_second": 564.59,
|
| 345 |
+
"eval_steps_per_second": 1.111,
|
| 346 |
+
"step": 2091
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 5.015631784628745,
|
| 350 |
+
"grad_norm": 23.5882625579834,
|
| 351 |
+
"learning_rate": 0.000105,
|
| 352 |
+
"loss": 3.5367,
|
| 353 |
+
"step": 2100
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 5.102475032566218,
|
| 357 |
+
"grad_norm": 20.266260147094727,
|
| 358 |
+
"learning_rate": 0.0001075,
|
| 359 |
+
"loss": 2.6965,
|
| 360 |
+
"step": 2150
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 5.189318280503691,
|
| 364 |
+
"grad_norm": 18.317874908447266,
|
| 365 |
+
"learning_rate": 0.00011,
|
| 366 |
+
"loss": 2.5427,
|
| 367 |
+
"step": 2200
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 5.276161528441164,
|
| 371 |
+
"grad_norm": 22.011568069458008,
|
| 372 |
+
"learning_rate": 0.00011250000000000001,
|
| 373 |
+
"loss": 2.4528,
|
| 374 |
+
"step": 2250
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 5.363004776378636,
|
| 378 |
+
"grad_norm": 21.177799224853516,
|
| 379 |
+
"learning_rate": 0.000115,
|
| 380 |
+
"loss": 2.3982,
|
| 381 |
+
"step": 2300
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 5.449848024316109,
|
| 385 |
+
"grad_norm": 20.553186416625977,
|
| 386 |
+
"learning_rate": 0.0001175,
|
| 387 |
+
"loss": 2.3483,
|
| 388 |
+
"step": 2350
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 5.536691272253583,
|
| 392 |
+
"grad_norm": 19.31122398376465,
|
| 393 |
+
"learning_rate": 0.00012,
|
| 394 |
+
"loss": 2.2953,
|
| 395 |
+
"step": 2400
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 5.623534520191055,
|
| 399 |
+
"grad_norm": 21.9901180267334,
|
| 400 |
+
"learning_rate": 0.0001225,
|
| 401 |
+
"loss": 2.2574,
|
| 402 |
+
"step": 2450
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 5.710377768128528,
|
| 406 |
+
"grad_norm": 19.90314292907715,
|
| 407 |
+
"learning_rate": 0.000125,
|
| 408 |
+
"loss": 2.2118,
|
| 409 |
+
"step": 2500
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 5.7972210160660005,
|
| 413 |
+
"grad_norm": 21.27450942993164,
|
| 414 |
+
"learning_rate": 0.0001275,
|
| 415 |
+
"loss": 2.175,
|
| 416 |
+
"step": 2550
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 5.884064264003474,
|
| 420 |
+
"grad_norm": 21.723154067993164,
|
| 421 |
+
"learning_rate": 0.00013000000000000002,
|
| 422 |
+
"loss": 2.1175,
|
| 423 |
+
"step": 2600
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 5.970907511940947,
|
| 427 |
+
"grad_norm": 19.191791534423828,
|
| 428 |
+
"learning_rate": 0.00013250000000000002,
|
| 429 |
+
"loss": 2.1001,
|
| 430 |
+
"step": 2650
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 5.998697351280938,
|
| 434 |
+
"eval_accuracy": 0.0,
|
| 435 |
+
"eval_loss": 4.229435443878174,
|
| 436 |
+
"eval_normalizer": 136004.0,
|
| 437 |
+
"eval_runtime": 102.8723,
|
| 438 |
+
"eval_samples_per_second": 573.04,
|
| 439 |
+
"eval_steps_per_second": 1.128,
|
| 440 |
+
"step": 2666
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 6.059053408597482,
|
| 444 |
+
"grad_norm": 15.839791297912598,
|
| 445 |
+
"learning_rate": 0.000135,
|
| 446 |
+
"loss": 1.9955,
|
| 447 |
+
"step": 2700
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 6.145896656534954,
|
| 451 |
+
"grad_norm": 16.21309471130371,
|
| 452 |
+
"learning_rate": 0.0001375,
|
| 453 |
+
"loss": 1.8325,
|
| 454 |
+
"step": 2750
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 6.232739904472427,
|
| 458 |
+
"grad_norm": 16.051525115966797,
|
| 459 |
+
"learning_rate": 0.00014000000000000001,
|
| 460 |
+
"loss": 1.7959,
|
| 461 |
+
"step": 2800
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 6.3195831524099,
|
| 465 |
+
"grad_norm": 15.809335708618164,
|
| 466 |
+
"learning_rate": 0.0001425,
|
| 467 |
+
"loss": 1.7435,
|
| 468 |
+
"step": 2850
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 6.406426400347373,
|
| 472 |
+
"grad_norm": 16.334001541137695,
|
| 473 |
+
"learning_rate": 0.000145,
|
| 474 |
+
"loss": 1.7093,
|
| 475 |
+
"step": 2900
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 6.493269648284846,
|
| 479 |
+
"grad_norm": 16.464365005493164,
|
| 480 |
+
"learning_rate": 0.0001475,
|
| 481 |
+
"loss": 1.694,
|
| 482 |
+
"step": 2950
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 6.580112896222319,
|
| 486 |
+
"grad_norm": 17.633712768554688,
|
| 487 |
+
"learning_rate": 0.00015,
|
| 488 |
+
"loss": 1.6506,
|
| 489 |
+
"step": 3000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 6.6669561441597915,
|
| 493 |
+
"grad_norm": 18.032367706298828,
|
| 494 |
+
"learning_rate": 0.0001525,
|
| 495 |
+
"loss": 1.6152,
|
| 496 |
+
"step": 3050
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 6.753799392097265,
|
| 500 |
+
"grad_norm": 15.881841659545898,
|
| 501 |
+
"learning_rate": 0.000155,
|
| 502 |
+
"loss": 1.5828,
|
| 503 |
+
"step": 3100
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 6.840642640034737,
|
| 507 |
+
"grad_norm": 17.653053283691406,
|
| 508 |
+
"learning_rate": 0.0001575,
|
| 509 |
+
"loss": 1.5578,
|
| 510 |
+
"step": 3150
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 6.92748588797221,
|
| 514 |
+
"grad_norm": 16.09905242919922,
|
| 515 |
+
"learning_rate": 0.00016,
|
| 516 |
+
"loss": 1.5391,
|
| 517 |
+
"step": 3200
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 6.998697351280938,
|
| 521 |
+
"eval_accuracy": 0.0,
|
| 522 |
+
"eval_loss": 4.745354652404785,
|
| 523 |
+
"eval_normalizer": 136349.0,
|
| 524 |
+
"eval_runtime": 103.5677,
|
| 525 |
+
"eval_samples_per_second": 569.193,
|
| 526 |
+
"eval_steps_per_second": 1.12,
|
| 527 |
+
"step": 3241
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 7.015631784628745,
|
| 531 |
+
"grad_norm": 16.049943923950195,
|
| 532 |
+
"learning_rate": 0.00016250000000000002,
|
| 533 |
+
"loss": 1.565,
|
| 534 |
+
"step": 3250
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 7.102475032566218,
|
| 538 |
+
"grad_norm": 15.44378662109375,
|
| 539 |
+
"learning_rate": 0.000165,
|
| 540 |
+
"loss": 1.6589,
|
| 541 |
+
"step": 3300
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 7.189318280503691,
|
| 545 |
+
"grad_norm": 14.940760612487793,
|
| 546 |
+
"learning_rate": 0.0001675,
|
| 547 |
+
"loss": 1.604,
|
| 548 |
+
"step": 3350
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 7.276161528441164,
|
| 552 |
+
"grad_norm": 15.526762008666992,
|
| 553 |
+
"learning_rate": 0.00017,
|
| 554 |
+
"loss": 1.5453,
|
| 555 |
+
"step": 3400
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 7.363004776378636,
|
| 559 |
+
"grad_norm": 15.957108497619629,
|
| 560 |
+
"learning_rate": 0.0001725,
|
| 561 |
+
"loss": 1.5156,
|
| 562 |
+
"step": 3450
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 7.449848024316109,
|
| 566 |
+
"grad_norm": 16.50444221496582,
|
| 567 |
+
"learning_rate": 0.000175,
|
| 568 |
+
"loss": 1.4938,
|
| 569 |
+
"step": 3500
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 7.536691272253583,
|
| 573 |
+
"grad_norm": 16.527673721313477,
|
| 574 |
+
"learning_rate": 0.0001775,
|
| 575 |
+
"loss": 1.4478,
|
| 576 |
+
"step": 3550
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 7.623534520191055,
|
| 580 |
+
"grad_norm": 15.978376388549805,
|
| 581 |
+
"learning_rate": 0.00017999999999999998,
|
| 582 |
+
"loss": 1.4137,
|
| 583 |
+
"step": 3600
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 7.710377768128528,
|
| 587 |
+
"grad_norm": 16.58029556274414,
|
| 588 |
+
"learning_rate": 0.0001825,
|
| 589 |
+
"loss": 1.397,
|
| 590 |
+
"step": 3650
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 7.7972210160660005,
|
| 594 |
+
"grad_norm": 17.977977752685547,
|
| 595 |
+
"learning_rate": 0.000185,
|
| 596 |
+
"loss": 1.3734,
|
| 597 |
+
"step": 3700
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 7.884064264003474,
|
| 601 |
+
"grad_norm": 16.021697998046875,
|
| 602 |
+
"learning_rate": 0.0001875,
|
| 603 |
+
"loss": 1.3308,
|
| 604 |
+
"step": 3750
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 7.970907511940947,
|
| 608 |
+
"grad_norm": 15.896733283996582,
|
| 609 |
+
"learning_rate": 0.00019,
|
| 610 |
+
"loss": 1.323,
|
| 611 |
+
"step": 3800
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 7.998697351280938,
|
| 615 |
+
"eval_accuracy": 0.0,
|
| 616 |
+
"eval_loss": 5.486119270324707,
|
| 617 |
+
"eval_normalizer": 136569.0,
|
| 618 |
+
"eval_runtime": 103.6313,
|
| 619 |
+
"eval_samples_per_second": 568.844,
|
| 620 |
+
"eval_steps_per_second": 1.119,
|
| 621 |
+
"step": 3816
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 8.059053408597482,
|
| 625 |
+
"grad_norm": 15.037135124206543,
|
| 626 |
+
"learning_rate": 0.00019250000000000002,
|
| 627 |
+
"loss": 1.4876,
|
| 628 |
+
"step": 3850
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 8.145896656534955,
|
| 632 |
+
"grad_norm": 15.31440544128418,
|
| 633 |
+
"learning_rate": 0.00019500000000000002,
|
| 634 |
+
"loss": 1.4557,
|
| 635 |
+
"step": 3900
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 8.232739904472428,
|
| 639 |
+
"grad_norm": 15.17057991027832,
|
| 640 |
+
"learning_rate": 0.0001975,
|
| 641 |
+
"loss": 1.4014,
|
| 642 |
+
"step": 3950
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 8.3195831524099,
|
| 646 |
+
"grad_norm": 16.278615951538086,
|
| 647 |
+
"learning_rate": 0.0002,
|
| 648 |
+
"loss": 1.3482,
|
| 649 |
+
"step": 4000
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 8.406426400347373,
|
| 653 |
+
"grad_norm": 14.423457145690918,
|
| 654 |
+
"learning_rate": 0.00020250000000000002,
|
| 655 |
+
"loss": 1.3199,
|
| 656 |
+
"step": 4050
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 8.493269648284846,
|
| 660 |
+
"grad_norm": 15.386080741882324,
|
| 661 |
+
"learning_rate": 0.000205,
|
| 662 |
+
"loss": 1.2817,
|
| 663 |
+
"step": 4100
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 8.58011289622232,
|
| 667 |
+
"grad_norm": 15.96497917175293,
|
| 668 |
+
"learning_rate": 0.0002075,
|
| 669 |
+
"loss": 1.2539,
|
| 670 |
+
"step": 4150
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 8.66695614415979,
|
| 674 |
+
"grad_norm": 14.696466445922852,
|
| 675 |
+
"learning_rate": 0.00021,
|
| 676 |
+
"loss": 1.2161,
|
| 677 |
+
"step": 4200
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 8.753799392097264,
|
| 681 |
+
"grad_norm": 15.68907642364502,
|
| 682 |
+
"learning_rate": 0.0002125,
|
| 683 |
+
"loss": 1.1881,
|
| 684 |
+
"step": 4250
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 8.840642640034737,
|
| 688 |
+
"grad_norm": 16.16935920715332,
|
| 689 |
+
"learning_rate": 0.000215,
|
| 690 |
+
"loss": 1.168,
|
| 691 |
+
"step": 4300
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 8.92748588797221,
|
| 695 |
+
"grad_norm": 17.611055374145508,
|
| 696 |
+
"learning_rate": 0.0002175,
|
| 697 |
+
"loss": 1.1338,
|
| 698 |
+
"step": 4350
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 8.998697351280939,
|
| 702 |
+
"eval_accuracy": 0.0,
|
| 703 |
+
"eval_loss": 6.356105804443359,
|
| 704 |
+
"eval_normalizer": 136839.0,
|
| 705 |
+
"eval_runtime": 104.8441,
|
| 706 |
+
"eval_samples_per_second": 562.264,
|
| 707 |
+
"eval_steps_per_second": 1.106,
|
| 708 |
+
"step": 4391
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 9.015631784628745,
|
| 712 |
+
"grad_norm": 22.64611053466797,
|
| 713 |
+
"learning_rate": 0.00022,
|
| 714 |
+
"loss": 1.1845,
|
| 715 |
+
"step": 4400
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 9.102475032566218,
|
| 719 |
+
"grad_norm": 14.135415077209473,
|
| 720 |
+
"learning_rate": 0.00022250000000000001,
|
| 721 |
+
"loss": 1.2931,
|
| 722 |
+
"step": 4450
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 9.189318280503691,
|
| 726 |
+
"grad_norm": 14.350916862487793,
|
| 727 |
+
"learning_rate": 0.00022500000000000002,
|
| 728 |
+
"loss": 1.195,
|
| 729 |
+
"step": 4500
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 9.276161528441163,
|
| 733 |
+
"grad_norm": 14.335700988769531,
|
| 734 |
+
"learning_rate": 0.0002275,
|
| 735 |
+
"loss": 1.1328,
|
| 736 |
+
"step": 4550
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 9.363004776378636,
|
| 740 |
+
"grad_norm": 13.810677528381348,
|
| 741 |
+
"learning_rate": 0.00023,
|
| 742 |
+
"loss": 1.0703,
|
| 743 |
+
"step": 4600
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 9.44984802431611,
|
| 747 |
+
"grad_norm": 14.181976318359375,
|
| 748 |
+
"learning_rate": 0.0002325,
|
| 749 |
+
"loss": 1.0392,
|
| 750 |
+
"step": 4650
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 9.536691272253583,
|
| 754 |
+
"grad_norm": 13.69727897644043,
|
| 755 |
+
"learning_rate": 0.000235,
|
| 756 |
+
"loss": 1.0131,
|
| 757 |
+
"step": 4700
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 9.623534520191056,
|
| 761 |
+
"grad_norm": 16.096031188964844,
|
| 762 |
+
"learning_rate": 0.0002375,
|
| 763 |
+
"loss": 0.9865,
|
| 764 |
+
"step": 4750
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 9.710377768128527,
|
| 768 |
+
"grad_norm": 15.79973316192627,
|
| 769 |
+
"learning_rate": 0.00024,
|
| 770 |
+
"loss": 0.9347,
|
| 771 |
+
"step": 4800
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 9.797221016066,
|
| 775 |
+
"grad_norm": 15.413460731506348,
|
| 776 |
+
"learning_rate": 0.00024249999999999999,
|
| 777 |
+
"loss": 0.9262,
|
| 778 |
+
"step": 4850
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 9.884064264003474,
|
| 782 |
+
"grad_norm": 16.058490753173828,
|
| 783 |
+
"learning_rate": 0.000245,
|
| 784 |
+
"loss": 0.8874,
|
| 785 |
+
"step": 4900
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 9.970907511940947,
|
| 789 |
+
"grad_norm": 14.666397094726562,
|
| 790 |
+
"learning_rate": 0.0002475,
|
| 791 |
+
"loss": 0.8514,
|
| 792 |
+
"step": 4950
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 9.998697351280939,
|
| 796 |
+
"eval_accuracy": 0.0,
|
| 797 |
+
"eval_loss": 7.307034969329834,
|
| 798 |
+
"eval_normalizer": 136494.0,
|
| 799 |
+
"eval_runtime": 103.6113,
|
| 800 |
+
"eval_samples_per_second": 568.954,
|
| 801 |
+
"eval_steps_per_second": 1.12,
|
| 802 |
+
"step": 4966
|
| 803 |
+
}
|
| 804 |
+
],
|
| 805 |
+
"logging_steps": 50,
|
| 806 |
+
"max_steps": 5750,
|
| 807 |
+
"num_input_tokens_seen": 0,
|
| 808 |
+
"num_train_epochs": 10,
|
| 809 |
+
"save_steps": 500,
|
| 810 |
+
"stateful_callbacks": {
|
| 811 |
+
"TrainerControl": {
|
| 812 |
+
"args": {
|
| 813 |
+
"should_epoch_stop": false,
|
| 814 |
+
"should_evaluate": false,
|
| 815 |
+
"should_log": false,
|
| 816 |
+
"should_save": true,
|
| 817 |
+
"should_training_stop": false
|
| 818 |
+
},
|
| 819 |
+
"attributes": {}
|
| 820 |
+
}
|
| 821 |
+
},
|
| 822 |
+
"total_flos": 4.628310664208394e+17,
|
| 823 |
+
"train_batch_size": 128,
|
| 824 |
+
"trial_name": null,
|
| 825 |
+
"trial_params": null
|
| 826 |
+
}
|
checkpoints/checkpoint-4966/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bdc08d0d14aa5507561d445c16c48b1cdaedb8d8bc2b2e27d8757638d2a8aee0
|
| 3 |
+
size 5432
|