Upload folder using huggingface_hub
Browse files- checkpoints/checkpoint-5220/config.json +26 -0
- checkpoints/checkpoint-5220/model.safetensors +3 -0
- checkpoints/checkpoint-5220/optimizer.pt +3 -0
- checkpoints/checkpoint-5220/rng_state.pth +3 -0
- checkpoints/checkpoint-5220/scheduler.pt +3 -0
- checkpoints/checkpoint-5220/trainer_state.json +861 -0
- checkpoints/checkpoint-5220/training_args.bin +3 -0
checkpoints/checkpoint-5220/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-5220/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5b7d6e46202343cb2fd27f7d1a2cd057aab3f4695ac27e4449985781b51236e0
|
| 3 |
+
size 504150808
|
checkpoints/checkpoint-5220/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7ed797ccb5307da2e088880d2fafc07167c05a19f39b0ad504669911c1c2d884
|
| 3 |
+
size 1008422138
|
checkpoints/checkpoint-5220/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:78d9e2eac83b9d6e59d13b8a6cf8e4727e03d733546392ea46345bed57305d38
|
| 3 |
+
size 14244
|
checkpoints/checkpoint-5220/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3c22e92b12373078f8b1a1a0411708ab510b99bc0b3dcf2614592865b8b0368a
|
| 3 |
+
size 1064
|
checkpoints/checkpoint-5220/trainer_state.json
ADDED
|
@@ -0,0 +1,861 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 9.998804113848362,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 5220,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.09567089213106912,
|
| 13 |
+
"grad_norm": 52.747867584228516,
|
| 14 |
+
"learning_rate": 2.5e-06,
|
| 15 |
+
"loss": 10.8165,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.19134178426213824,
|
| 20 |
+
"grad_norm": 66.05134582519531,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 10.054,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.2870126763932074,
|
| 27 |
+
"grad_norm": 53.76071548461914,
|
| 28 |
+
"learning_rate": 7.5e-06,
|
| 29 |
+
"loss": 9.0799,
|
| 30 |
+
"step": 150
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.3826835685242765,
|
| 34 |
+
"grad_norm": 37.95576477050781,
|
| 35 |
+
"learning_rate": 1e-05,
|
| 36 |
+
"loss": 9.0591,
|
| 37 |
+
"step": 200
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.4783544606553456,
|
| 41 |
+
"grad_norm": 44.30142593383789,
|
| 42 |
+
"learning_rate": 1.25e-05,
|
| 43 |
+
"loss": 9.0954,
|
| 44 |
+
"step": 250
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.5740253527864148,
|
| 48 |
+
"grad_norm": 44.76091766357422,
|
| 49 |
+
"learning_rate": 1.5e-05,
|
| 50 |
+
"loss": 8.2617,
|
| 51 |
+
"step": 300
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.6696962449174838,
|
| 55 |
+
"grad_norm": 36.74516677856445,
|
| 56 |
+
"learning_rate": 1.7500000000000002e-05,
|
| 57 |
+
"loss": 7.6069,
|
| 58 |
+
"step": 350
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.765367137048553,
|
| 62 |
+
"grad_norm": 42.63846206665039,
|
| 63 |
+
"learning_rate": 2e-05,
|
| 64 |
+
"loss": 7.278,
|
| 65 |
+
"step": 400
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.8610380291796221,
|
| 69 |
+
"grad_norm": 26.08951759338379,
|
| 70 |
+
"learning_rate": 2.2499999999999998e-05,
|
| 71 |
+
"loss": 7.0434,
|
| 72 |
+
"step": 450
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.9567089213106912,
|
| 76 |
+
"grad_norm": 23.093669891357422,
|
| 77 |
+
"learning_rate": 2.5e-05,
|
| 78 |
+
"loss": 7.5574,
|
| 79 |
+
"step": 500
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.9988041138483617,
|
| 83 |
+
"eval_accuracy": 0.0,
|
| 84 |
+
"eval_loss": 7.002418041229248,
|
| 85 |
+
"eval_normalizer": 94012.0,
|
| 86 |
+
"eval_runtime": 93.6448,
|
| 87 |
+
"eval_samples_per_second": 570.849,
|
| 88 |
+
"eval_steps_per_second": 1.121,
|
| 89 |
+
"step": 522
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.0535756995933987,
|
| 93 |
+
"grad_norm": 21.285545349121094,
|
| 94 |
+
"learning_rate": 2.75e-05,
|
| 95 |
+
"loss": 6.9275,
|
| 96 |
+
"step": 550
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.1492465917244679,
|
| 100 |
+
"grad_norm": 33.146881103515625,
|
| 101 |
+
"learning_rate": 3e-05,
|
| 102 |
+
"loss": 6.8179,
|
| 103 |
+
"step": 600
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 1.244917483855537,
|
| 107 |
+
"grad_norm": 28.52762222290039,
|
| 108 |
+
"learning_rate": 3.2500000000000004e-05,
|
| 109 |
+
"loss": 5.2322,
|
| 110 |
+
"step": 650
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 1.3405883759866062,
|
| 114 |
+
"grad_norm": 33.8976936340332,
|
| 115 |
+
"learning_rate": 3.5000000000000004e-05,
|
| 116 |
+
"loss": 5.0881,
|
| 117 |
+
"step": 700
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 1.4362592681176751,
|
| 121 |
+
"grad_norm": 31.849340438842773,
|
| 122 |
+
"learning_rate": 3.75e-05,
|
| 123 |
+
"loss": 6.3516,
|
| 124 |
+
"step": 750
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 1.5319301602487443,
|
| 128 |
+
"grad_norm": 26.819744110107422,
|
| 129 |
+
"learning_rate": 4e-05,
|
| 130 |
+
"loss": 5.6243,
|
| 131 |
+
"step": 800
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 1.6276010523798135,
|
| 135 |
+
"grad_norm": 27.690715789794922,
|
| 136 |
+
"learning_rate": 4.25e-05,
|
| 137 |
+
"loss": 5.4867,
|
| 138 |
+
"step": 850
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 1.7232719445108826,
|
| 142 |
+
"grad_norm": 39.06813430786133,
|
| 143 |
+
"learning_rate": 4.4999999999999996e-05,
|
| 144 |
+
"loss": 5.3091,
|
| 145 |
+
"step": 900
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 1.8189428366419516,
|
| 149 |
+
"grad_norm": 25.289762496948242,
|
| 150 |
+
"learning_rate": 4.75e-05,
|
| 151 |
+
"loss": 5.9413,
|
| 152 |
+
"step": 950
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 1.9146137287730207,
|
| 156 |
+
"grad_norm": 31.576900482177734,
|
| 157 |
+
"learning_rate": 5e-05,
|
| 158 |
+
"loss": 6.3911,
|
| 159 |
+
"step": 1000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.9988041138483617,
|
| 163 |
+
"eval_accuracy": 0.0,
|
| 164 |
+
"eval_loss": 6.082557201385498,
|
| 165 |
+
"eval_normalizer": 93558.0,
|
| 166 |
+
"eval_runtime": 86.1348,
|
| 167 |
+
"eval_samples_per_second": 620.62,
|
| 168 |
+
"eval_steps_per_second": 1.219,
|
| 169 |
+
"step": 1044
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 2.0114805070557282,
|
| 173 |
+
"grad_norm": 36.15195846557617,
|
| 174 |
+
"learning_rate": 5.25e-05,
|
| 175 |
+
"loss": 6.4163,
|
| 176 |
+
"step": 1050
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 2.1071513991867974,
|
| 180 |
+
"grad_norm": 28.969654083251953,
|
| 181 |
+
"learning_rate": 5.5e-05,
|
| 182 |
+
"loss": 5.9146,
|
| 183 |
+
"step": 1100
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 2.2028222913178666,
|
| 187 |
+
"grad_norm": 31.072589874267578,
|
| 188 |
+
"learning_rate": 5.75e-05,
|
| 189 |
+
"loss": 4.9223,
|
| 190 |
+
"step": 1150
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 2.2984931834489357,
|
| 194 |
+
"grad_norm": 33.17365646362305,
|
| 195 |
+
"learning_rate": 6e-05,
|
| 196 |
+
"loss": 3.9321,
|
| 197 |
+
"step": 1200
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 2.394164075580005,
|
| 201 |
+
"grad_norm": 20.012985229492188,
|
| 202 |
+
"learning_rate": 6.25e-05,
|
| 203 |
+
"loss": 5.1453,
|
| 204 |
+
"step": 1250
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 2.489834967711074,
|
| 208 |
+
"grad_norm": 33.7502326965332,
|
| 209 |
+
"learning_rate": 6.500000000000001e-05,
|
| 210 |
+
"loss": 4.8,
|
| 211 |
+
"step": 1300
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 2.5855058598421428,
|
| 215 |
+
"grad_norm": 32.18284225463867,
|
| 216 |
+
"learning_rate": 6.75e-05,
|
| 217 |
+
"loss": 4.5708,
|
| 218 |
+
"step": 1350
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.6811767519732124,
|
| 222 |
+
"grad_norm": 26.967090606689453,
|
| 223 |
+
"learning_rate": 7.000000000000001e-05,
|
| 224 |
+
"loss": 4.479,
|
| 225 |
+
"step": 1400
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 2.776847644104281,
|
| 229 |
+
"grad_norm": 21.180952072143555,
|
| 230 |
+
"learning_rate": 7.25e-05,
|
| 231 |
+
"loss": 4.847,
|
| 232 |
+
"step": 1450
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 2.8725185362353503,
|
| 236 |
+
"grad_norm": 34.09960174560547,
|
| 237 |
+
"learning_rate": 7.5e-05,
|
| 238 |
+
"loss": 5.2215,
|
| 239 |
+
"step": 1500
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 2.9681894283664194,
|
| 243 |
+
"grad_norm": 30.934101104736328,
|
| 244 |
+
"learning_rate": 7.75e-05,
|
| 245 |
+
"loss": 6.1036,
|
| 246 |
+
"step": 1550
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 2.9988041138483617,
|
| 250 |
+
"eval_accuracy": 0.0,
|
| 251 |
+
"eval_loss": 5.322023868560791,
|
| 252 |
+
"eval_normalizer": 94216.0,
|
| 253 |
+
"eval_runtime": 101.6167,
|
| 254 |
+
"eval_samples_per_second": 526.065,
|
| 255 |
+
"eval_steps_per_second": 1.033,
|
| 256 |
+
"step": 1566
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 3.065056206649127,
|
| 260 |
+
"grad_norm": 25.038644790649414,
|
| 261 |
+
"learning_rate": 8e-05,
|
| 262 |
+
"loss": 5.2087,
|
| 263 |
+
"step": 1600
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 3.160727098780196,
|
| 267 |
+
"grad_norm": 32.0327033996582,
|
| 268 |
+
"learning_rate": 8.25e-05,
|
| 269 |
+
"loss": 5.0918,
|
| 270 |
+
"step": 1650
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 3.2563979909112652,
|
| 274 |
+
"grad_norm": 33.0177001953125,
|
| 275 |
+
"learning_rate": 8.5e-05,
|
| 276 |
+
"loss": 3.4156,
|
| 277 |
+
"step": 1700
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 3.3520688830423344,
|
| 281 |
+
"grad_norm": 24.040756225585938,
|
| 282 |
+
"learning_rate": 8.75e-05,
|
| 283 |
+
"loss": 3.7571,
|
| 284 |
+
"step": 1750
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 3.4477397751734036,
|
| 288 |
+
"grad_norm": 27.80860710144043,
|
| 289 |
+
"learning_rate": 8.999999999999999e-05,
|
| 290 |
+
"loss": 4.6411,
|
| 291 |
+
"step": 1800
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 3.5434106673044727,
|
| 295 |
+
"grad_norm": 25.351137161254883,
|
| 296 |
+
"learning_rate": 9.25e-05,
|
| 297 |
+
"loss": 4.0891,
|
| 298 |
+
"step": 1850
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 3.639081559435542,
|
| 302 |
+
"grad_norm": 25.684823989868164,
|
| 303 |
+
"learning_rate": 9.5e-05,
|
| 304 |
+
"loss": 4.0587,
|
| 305 |
+
"step": 1900
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 3.734752451566611,
|
| 309 |
+
"grad_norm": 24.86536407470703,
|
| 310 |
+
"learning_rate": 9.750000000000001e-05,
|
| 311 |
+
"loss": 4.0672,
|
| 312 |
+
"step": 1950
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 3.83042334369768,
|
| 316 |
+
"grad_norm": 23.63920021057129,
|
| 317 |
+
"learning_rate": 0.0001,
|
| 318 |
+
"loss": 4.7268,
|
| 319 |
+
"step": 2000
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 3.926094235828749,
|
| 323 |
+
"grad_norm": 25.422239303588867,
|
| 324 |
+
"learning_rate": 0.0001025,
|
| 325 |
+
"loss": 5.369,
|
| 326 |
+
"step": 2050
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 3.9988041138483617,
|
| 330 |
+
"eval_accuracy": 0.0,
|
| 331 |
+
"eval_loss": 4.845930099487305,
|
| 332 |
+
"eval_normalizer": 93665.0,
|
| 333 |
+
"eval_runtime": 98.1242,
|
| 334 |
+
"eval_samples_per_second": 544.789,
|
| 335 |
+
"eval_steps_per_second": 1.07,
|
| 336 |
+
"step": 2088
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 4.0229610141114565,
|
| 340 |
+
"grad_norm": 28.550003051757812,
|
| 341 |
+
"learning_rate": 0.000105,
|
| 342 |
+
"loss": 5.131,
|
| 343 |
+
"step": 2100
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 4.118631906242526,
|
| 347 |
+
"grad_norm": 24.855363845825195,
|
| 348 |
+
"learning_rate": 0.0001075,
|
| 349 |
+
"loss": 4.85,
|
| 350 |
+
"step": 2150
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 4.214302798373595,
|
| 354 |
+
"grad_norm": 28.85999870300293,
|
| 355 |
+
"learning_rate": 0.00011,
|
| 356 |
+
"loss": 3.6755,
|
| 357 |
+
"step": 2200
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 4.309973690504664,
|
| 361 |
+
"grad_norm": 26.07246971130371,
|
| 362 |
+
"learning_rate": 0.00011250000000000001,
|
| 363 |
+
"loss": 3.0516,
|
| 364 |
+
"step": 2250
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 4.405644582635733,
|
| 368 |
+
"grad_norm": 66.89540100097656,
|
| 369 |
+
"learning_rate": 0.000115,
|
| 370 |
+
"loss": 4.28,
|
| 371 |
+
"step": 2300
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 4.501315474766802,
|
| 375 |
+
"grad_norm": 23.402463912963867,
|
| 376 |
+
"learning_rate": 0.0001175,
|
| 377 |
+
"loss": 3.8331,
|
| 378 |
+
"step": 2350
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 4.596986366897871,
|
| 382 |
+
"grad_norm": 24.565364837646484,
|
| 383 |
+
"learning_rate": 0.00012,
|
| 384 |
+
"loss": 3.8233,
|
| 385 |
+
"step": 2400
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 4.69265725902894,
|
| 389 |
+
"grad_norm": 23.60770034790039,
|
| 390 |
+
"learning_rate": 0.0001225,
|
| 391 |
+
"loss": 3.7483,
|
| 392 |
+
"step": 2450
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 4.78832815116001,
|
| 396 |
+
"grad_norm": 19.10536003112793,
|
| 397 |
+
"learning_rate": 0.000125,
|
| 398 |
+
"loss": 4.1922,
|
| 399 |
+
"step": 2500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 4.8839990432910785,
|
| 403 |
+
"grad_norm": 23.9981746673584,
|
| 404 |
+
"learning_rate": 0.0001275,
|
| 405 |
+
"loss": 4.5636,
|
| 406 |
+
"step": 2550
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 4.979669935422148,
|
| 410 |
+
"grad_norm": 52.10109329223633,
|
| 411 |
+
"learning_rate": 0.00013000000000000002,
|
| 412 |
+
"loss": 5.1324,
|
| 413 |
+
"step": 2600
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 4.998804113848362,
|
| 417 |
+
"eval_accuracy": 0.0,
|
| 418 |
+
"eval_loss": 4.496237754821777,
|
| 419 |
+
"eval_normalizer": 94313.0,
|
| 420 |
+
"eval_runtime": 86.4844,
|
| 421 |
+
"eval_samples_per_second": 618.112,
|
| 422 |
+
"eval_steps_per_second": 1.214,
|
| 423 |
+
"step": 2610
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 5.076536713704855,
|
| 427 |
+
"grad_norm": 23.186723709106445,
|
| 428 |
+
"learning_rate": 0.00013250000000000002,
|
| 429 |
+
"loss": 4.5323,
|
| 430 |
+
"step": 2650
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 5.172207605835925,
|
| 434 |
+
"grad_norm": 28.158355712890625,
|
| 435 |
+
"learning_rate": 0.000135,
|
| 436 |
+
"loss": 4.2118,
|
| 437 |
+
"step": 2700
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 5.2678784979669935,
|
| 441 |
+
"grad_norm": 28.501588821411133,
|
| 442 |
+
"learning_rate": 0.0001375,
|
| 443 |
+
"loss": 2.9381,
|
| 444 |
+
"step": 2750
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 5.363549390098063,
|
| 448 |
+
"grad_norm": 18.10572052001953,
|
| 449 |
+
"learning_rate": 0.00014000000000000001,
|
| 450 |
+
"loss": 3.4088,
|
| 451 |
+
"step": 2800
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 5.459220282229132,
|
| 455 |
+
"grad_norm": 26.142528533935547,
|
| 456 |
+
"learning_rate": 0.0001425,
|
| 457 |
+
"loss": 3.8884,
|
| 458 |
+
"step": 2850
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 5.5548911743602005,
|
| 462 |
+
"grad_norm": 22.223371505737305,
|
| 463 |
+
"learning_rate": 0.000145,
|
| 464 |
+
"loss": 3.6124,
|
| 465 |
+
"step": 2900
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 5.65056206649127,
|
| 469 |
+
"grad_norm": 24.494014739990234,
|
| 470 |
+
"learning_rate": 0.0001475,
|
| 471 |
+
"loss": 3.6075,
|
| 472 |
+
"step": 2950
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 5.746232958622339,
|
| 476 |
+
"grad_norm": 21.019962310791016,
|
| 477 |
+
"learning_rate": 0.00015,
|
| 478 |
+
"loss": 3.6963,
|
| 479 |
+
"step": 3000
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 5.8419038507534085,
|
| 483 |
+
"grad_norm": 20.909692764282227,
|
| 484 |
+
"learning_rate": 0.0001525,
|
| 485 |
+
"loss": 4.1607,
|
| 486 |
+
"step": 3050
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 5.937574742884477,
|
| 490 |
+
"grad_norm": 27.56682586669922,
|
| 491 |
+
"learning_rate": 0.000155,
|
| 492 |
+
"loss": 4.8602,
|
| 493 |
+
"step": 3100
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 5.998804113848362,
|
| 497 |
+
"eval_accuracy": 0.0,
|
| 498 |
+
"eval_loss": 4.2728495597839355,
|
| 499 |
+
"eval_normalizer": 93901.0,
|
| 500 |
+
"eval_runtime": 81.5686,
|
| 501 |
+
"eval_samples_per_second": 655.363,
|
| 502 |
+
"eval_steps_per_second": 1.287,
|
| 503 |
+
"step": 3132
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 6.034441521167185,
|
| 507 |
+
"grad_norm": 24.819232940673828,
|
| 508 |
+
"learning_rate": 0.0001575,
|
| 509 |
+
"loss": 4.4373,
|
| 510 |
+
"step": 3150
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 6.130112413298254,
|
| 514 |
+
"grad_norm": 24.921003341674805,
|
| 515 |
+
"learning_rate": 0.00016,
|
| 516 |
+
"loss": 4.3447,
|
| 517 |
+
"step": 3200
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 6.2257833054293235,
|
| 521 |
+
"grad_norm": 25.00433349609375,
|
| 522 |
+
"learning_rate": 0.00016250000000000002,
|
| 523 |
+
"loss": 3.0774,
|
| 524 |
+
"step": 3250
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 6.321454197560392,
|
| 528 |
+
"grad_norm": 26.17780113220215,
|
| 529 |
+
"learning_rate": 0.000165,
|
| 530 |
+
"loss": 2.7673,
|
| 531 |
+
"step": 3300
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 6.417125089691462,
|
| 535 |
+
"grad_norm": 30.692371368408203,
|
| 536 |
+
"learning_rate": 0.0001675,
|
| 537 |
+
"loss": 3.8763,
|
| 538 |
+
"step": 3350
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 6.5127959818225305,
|
| 542 |
+
"grad_norm": 24.009729385375977,
|
| 543 |
+
"learning_rate": 0.00017,
|
| 544 |
+
"loss": 3.4603,
|
| 545 |
+
"step": 3400
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 6.608466873953599,
|
| 549 |
+
"grad_norm": 25.29176902770996,
|
| 550 |
+
"learning_rate": 0.0001725,
|
| 551 |
+
"loss": 3.4791,
|
| 552 |
+
"step": 3450
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 6.704137766084669,
|
| 556 |
+
"grad_norm": 25.014236450195312,
|
| 557 |
+
"learning_rate": 0.000175,
|
| 558 |
+
"loss": 3.4272,
|
| 559 |
+
"step": 3500
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 6.7998086582157375,
|
| 563 |
+
"grad_norm": 18.21346092224121,
|
| 564 |
+
"learning_rate": 0.0001775,
|
| 565 |
+
"loss": 3.8766,
|
| 566 |
+
"step": 3550
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 6.895479550346807,
|
| 570 |
+
"grad_norm": 22.74988555908203,
|
| 571 |
+
"learning_rate": 0.00017999999999999998,
|
| 572 |
+
"loss": 4.2445,
|
| 573 |
+
"step": 3600
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 6.991150442477876,
|
| 577 |
+
"grad_norm": 34.44709777832031,
|
| 578 |
+
"learning_rate": 0.0001825,
|
| 579 |
+
"loss": 4.554,
|
| 580 |
+
"step": 3650
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 6.998804113848362,
|
| 584 |
+
"eval_accuracy": 0.0,
|
| 585 |
+
"eval_loss": 4.073133945465088,
|
| 586 |
+
"eval_normalizer": 94462.0,
|
| 587 |
+
"eval_runtime": 86.9375,
|
| 588 |
+
"eval_samples_per_second": 614.89,
|
| 589 |
+
"eval_steps_per_second": 1.208,
|
| 590 |
+
"step": 3654
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 7.088017220760584,
|
| 594 |
+
"grad_norm": 22.7972354888916,
|
| 595 |
+
"learning_rate": 0.000185,
|
| 596 |
+
"loss": 4.0644,
|
| 597 |
+
"step": 3700
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 7.1836881128916525,
|
| 601 |
+
"grad_norm": 32.11361312866211,
|
| 602 |
+
"learning_rate": 0.0001875,
|
| 603 |
+
"loss": 3.6507,
|
| 604 |
+
"step": 3750
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 7.279359005022722,
|
| 608 |
+
"grad_norm": 23.8944034576416,
|
| 609 |
+
"learning_rate": 0.00019,
|
| 610 |
+
"loss": 2.7012,
|
| 611 |
+
"step": 3800
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 7.375029897153791,
|
| 615 |
+
"grad_norm": 16.63042449951172,
|
| 616 |
+
"learning_rate": 0.00019250000000000002,
|
| 617 |
+
"loss": 3.2761,
|
| 618 |
+
"step": 3850
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 7.4707007892848605,
|
| 622 |
+
"grad_norm": 24.039871215820312,
|
| 623 |
+
"learning_rate": 0.00019500000000000002,
|
| 624 |
+
"loss": 3.5065,
|
| 625 |
+
"step": 3900
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 7.566371681415929,
|
| 629 |
+
"grad_norm": 22.99151039123535,
|
| 630 |
+
"learning_rate": 0.0001975,
|
| 631 |
+
"loss": 3.362,
|
| 632 |
+
"step": 3950
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 7.662042573546998,
|
| 636 |
+
"grad_norm": 24.13475227355957,
|
| 637 |
+
"learning_rate": 0.0002,
|
| 638 |
+
"loss": 3.3509,
|
| 639 |
+
"step": 4000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 7.7577134656780675,
|
| 643 |
+
"grad_norm": 24.082887649536133,
|
| 644 |
+
"learning_rate": 0.00020250000000000002,
|
| 645 |
+
"loss": 3.5216,
|
| 646 |
+
"step": 4050
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 7.853384357809136,
|
| 650 |
+
"grad_norm": 19.067625045776367,
|
| 651 |
+
"learning_rate": 0.000205,
|
| 652 |
+
"loss": 3.8183,
|
| 653 |
+
"step": 4100
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 7.949055249940206,
|
| 657 |
+
"grad_norm": 21.847949981689453,
|
| 658 |
+
"learning_rate": 0.0002075,
|
| 659 |
+
"loss": 4.5382,
|
| 660 |
+
"step": 4150
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 7.998804113848362,
|
| 664 |
+
"eval_accuracy": 0.0,
|
| 665 |
+
"eval_loss": 3.880175828933716,
|
| 666 |
+
"eval_normalizer": 93833.0,
|
| 667 |
+
"eval_runtime": 91.0937,
|
| 668 |
+
"eval_samples_per_second": 586.835,
|
| 669 |
+
"eval_steps_per_second": 1.153,
|
| 670 |
+
"step": 4176
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 8.045922028222913,
|
| 674 |
+
"grad_norm": 22.053064346313477,
|
| 675 |
+
"learning_rate": 0.00021,
|
| 676 |
+
"loss": 4.0127,
|
| 677 |
+
"step": 4200
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 8.141592920353983,
|
| 681 |
+
"grad_norm": 23.286514282226562,
|
| 682 |
+
"learning_rate": 0.0002125,
|
| 683 |
+
"loss": 4.067,
|
| 684 |
+
"step": 4250
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 8.237263812485052,
|
| 688 |
+
"grad_norm": 24.23642921447754,
|
| 689 |
+
"learning_rate": 0.000215,
|
| 690 |
+
"loss": 2.7148,
|
| 691 |
+
"step": 4300
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 8.33293470461612,
|
| 695 |
+
"grad_norm": 34.11505126953125,
|
| 696 |
+
"learning_rate": 0.0002175,
|
| 697 |
+
"loss": 2.6508,
|
| 698 |
+
"step": 4350
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 8.42860559674719,
|
| 702 |
+
"grad_norm": 24.227184295654297,
|
| 703 |
+
"learning_rate": 0.00022,
|
| 704 |
+
"loss": 3.6736,
|
| 705 |
+
"step": 4400
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 8.52427648887826,
|
| 709 |
+
"grad_norm": 26.133562088012695,
|
| 710 |
+
"learning_rate": 0.00022250000000000001,
|
| 711 |
+
"loss": 3.2629,
|
| 712 |
+
"step": 4450
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 8.619947381009329,
|
| 716 |
+
"grad_norm": 23.79113006591797,
|
| 717 |
+
"learning_rate": 0.00022500000000000002,
|
| 718 |
+
"loss": 3.2882,
|
| 719 |
+
"step": 4500
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 8.715618273140397,
|
| 723 |
+
"grad_norm": 24.995574951171875,
|
| 724 |
+
"learning_rate": 0.0002275,
|
| 725 |
+
"loss": 3.2572,
|
| 726 |
+
"step": 4550
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 8.811289165271466,
|
| 730 |
+
"grad_norm": 20.96610450744629,
|
| 731 |
+
"learning_rate": 0.00023,
|
| 732 |
+
"loss": 3.7017,
|
| 733 |
+
"step": 4600
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 8.906960057402536,
|
| 737 |
+
"grad_norm": 20.328344345092773,
|
| 738 |
+
"learning_rate": 0.0002325,
|
| 739 |
+
"loss": 4.062,
|
| 740 |
+
"step": 4650
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 8.998804113848362,
|
| 744 |
+
"eval_accuracy": 0.0,
|
| 745 |
+
"eval_loss": 3.809931755065918,
|
| 746 |
+
"eval_normalizer": 93564.0,
|
| 747 |
+
"eval_runtime": 87.1052,
|
| 748 |
+
"eval_samples_per_second": 613.706,
|
| 749 |
+
"eval_steps_per_second": 1.205,
|
| 750 |
+
"step": 4698
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 9.003826835685242,
|
| 754 |
+
"grad_norm": 30.114665985107422,
|
| 755 |
+
"learning_rate": 0.000235,
|
| 756 |
+
"loss": 4.2445,
|
| 757 |
+
"step": 4700
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 9.099497727816312,
|
| 761 |
+
"grad_norm": 23.41254234313965,
|
| 762 |
+
"learning_rate": 0.0002375,
|
| 763 |
+
"loss": 3.7704,
|
| 764 |
+
"step": 4750
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 9.195168619947381,
|
| 768 |
+
"grad_norm": 24.50333023071289,
|
| 769 |
+
"learning_rate": 0.00024,
|
| 770 |
+
"loss": 3.2782,
|
| 771 |
+
"step": 4800
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 9.29083951207845,
|
| 775 |
+
"grad_norm": 24.675151824951172,
|
| 776 |
+
"learning_rate": 0.00024249999999999999,
|
| 777 |
+
"loss": 2.5975,
|
| 778 |
+
"step": 4850
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 9.386510404209519,
|
| 782 |
+
"grad_norm": 15.504040718078613,
|
| 783 |
+
"learning_rate": 0.000245,
|
| 784 |
+
"loss": 3.1972,
|
| 785 |
+
"step": 4900
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 9.482181296340588,
|
| 789 |
+
"grad_norm": 24.456554412841797,
|
| 790 |
+
"learning_rate": 0.0002475,
|
| 791 |
+
"loss": 3.2838,
|
| 792 |
+
"step": 4950
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 9.577852188471658,
|
| 796 |
+
"grad_norm": 24.348533630371094,
|
| 797 |
+
"learning_rate": 0.00025,
|
| 798 |
+
"loss": 3.2218,
|
| 799 |
+
"step": 5000
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 9.673523080602727,
|
| 803 |
+
"grad_norm": 23.819198608398438,
|
| 804 |
+
"learning_rate": 0.0002525,
|
| 805 |
+
"loss": 3.2363,
|
| 806 |
+
"step": 5050
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 9.769193972733795,
|
| 810 |
+
"grad_norm": 23.072057723999023,
|
| 811 |
+
"learning_rate": 0.000255,
|
| 812 |
+
"loss": 3.4138,
|
| 813 |
+
"step": 5100
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 9.864864864864865,
|
| 817 |
+
"grad_norm": 22.32210350036621,
|
| 818 |
+
"learning_rate": 0.0002575,
|
| 819 |
+
"loss": 3.6108,
|
| 820 |
+
"step": 5150
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 9.960535756995935,
|
| 824 |
+
"grad_norm": 22.900392532348633,
|
| 825 |
+
"learning_rate": 0.00026000000000000003,
|
| 826 |
+
"loss": 4.3926,
|
| 827 |
+
"step": 5200
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 9.998804113848362,
|
| 831 |
+
"eval_accuracy": 0.0,
|
| 832 |
+
"eval_loss": 3.749897003173828,
|
| 833 |
+
"eval_normalizer": 93774.0,
|
| 834 |
+
"eval_runtime": 87.6887,
|
| 835 |
+
"eval_samples_per_second": 609.622,
|
| 836 |
+
"eval_steps_per_second": 1.197,
|
| 837 |
+
"step": 5220
|
| 838 |
+
}
|
| 839 |
+
],
|
| 840 |
+
"logging_steps": 50,
|
| 841 |
+
"max_steps": 5220,
|
| 842 |
+
"num_input_tokens_seen": 0,
|
| 843 |
+
"num_train_epochs": 10,
|
| 844 |
+
"save_steps": 500,
|
| 845 |
+
"stateful_callbacks": {
|
| 846 |
+
"TrainerControl": {
|
| 847 |
+
"args": {
|
| 848 |
+
"should_epoch_stop": false,
|
| 849 |
+
"should_evaluate": false,
|
| 850 |
+
"should_log": false,
|
| 851 |
+
"should_save": true,
|
| 852 |
+
"should_training_stop": true
|
| 853 |
+
},
|
| 854 |
+
"attributes": {}
|
| 855 |
+
}
|
| 856 |
+
},
|
| 857 |
+
"total_flos": 5.786778914249441e+17,
|
| 858 |
+
"train_batch_size": 128,
|
| 859 |
+
"trial_name": null,
|
| 860 |
+
"trial_params": null
|
| 861 |
+
}
|
checkpoints/checkpoint-5220/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1dba3751aa7691a086bca0b3920013c2e5e1509ccf2a749a7547f9bd962cdd0d
|
| 3 |
+
size 5432
|