Training in progress, step 23500, checkpoint
Browse files
checkpoint-23500/config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "openai/clip-vit-large-patch14-336",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"CLIPModel"
|
| 5 |
+
],
|
| 6 |
+
"initializer_factor": 1.0,
|
| 7 |
+
"logit_scale_init_value": 2.6592,
|
| 8 |
+
"model_type": "clip",
|
| 9 |
+
"projection_dim": 768,
|
| 10 |
+
"text_config": {
|
| 11 |
+
"dropout": 0.0,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"model_type": "clip_text_model",
|
| 15 |
+
"num_attention_heads": 12,
|
| 16 |
+
"projection_dim": 768
|
| 17 |
+
},
|
| 18 |
+
"torch_dtype": "float32",
|
| 19 |
+
"transformers_version": "4.45.0.dev0",
|
| 20 |
+
"vision_config": {
|
| 21 |
+
"dropout": 0.0,
|
| 22 |
+
"hidden_size": 1024,
|
| 23 |
+
"image_size": 336,
|
| 24 |
+
"intermediate_size": 4096,
|
| 25 |
+
"model_type": "clip_vision_model",
|
| 26 |
+
"num_attention_heads": 16,
|
| 27 |
+
"num_hidden_layers": 24,
|
| 28 |
+
"patch_size": 14,
|
| 29 |
+
"projection_dim": 768
|
| 30 |
+
}
|
| 31 |
+
}
|
checkpoint-23500/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bfb38bd70666d3f71d0fde8fbd50335ad87c2111db8f1657011f7a0ab321e083
|
| 3 |
+
size 1711848436
|
checkpoint-23500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:64c35a0d2b928aca7353b95d61a97faa2ad49367310e5b789252bbc264f421d4
|
| 3 |
+
size 3424043887
|
checkpoint-23500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c4e011e4da26830e432615c30ab10725bdcb4903c39c3eefb707c88f12b2d9cb
|
| 3 |
+
size 14503
|
checkpoint-23500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4618fdb7e20f4ebcce6e2afd850ceeffc57e2487d063caef4907a2c0410e0fe3
|
| 3 |
+
size 623
|
checkpoint-23500/trainer_state.json
ADDED
|
@@ -0,0 +1,738 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.17657655477523804,
|
| 3 |
+
"best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e4l57-l/checkpoint-23500",
|
| 4 |
+
"epoch": 3.935689164294088,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 23500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.09211495946941783,
|
| 13 |
+
"grad_norm": 3.0293657779693604,
|
| 14 |
+
"learning_rate": 4.884856300663227e-07,
|
| 15 |
+
"loss": 0.271,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.09211495946941783,
|
| 20 |
+
"eval_loss": 1.069305419921875,
|
| 21 |
+
"eval_runtime": 75.1409,
|
| 22 |
+
"eval_samples_per_second": 16.063,
|
| 23 |
+
"eval_steps_per_second": 2.01,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.18422991893883567,
|
| 28 |
+
"grad_norm": 7.3040852546691895,
|
| 29 |
+
"learning_rate": 4.769712601326456e-07,
|
| 30 |
+
"loss": 0.2493,
|
| 31 |
+
"step": 1000
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.18422991893883567,
|
| 35 |
+
"eval_loss": 0.9427077770233154,
|
| 36 |
+
"eval_runtime": 74.8926,
|
| 37 |
+
"eval_samples_per_second": 16.116,
|
| 38 |
+
"eval_steps_per_second": 2.016,
|
| 39 |
+
"step": 1000
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.2763448784082535,
|
| 43 |
+
"grad_norm": 0.020736657083034515,
|
| 44 |
+
"learning_rate": 4.654568901989683e-07,
|
| 45 |
+
"loss": 0.2348,
|
| 46 |
+
"step": 1500
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.2763448784082535,
|
| 50 |
+
"eval_loss": 0.8726764917373657,
|
| 51 |
+
"eval_runtime": 76.6621,
|
| 52 |
+
"eval_samples_per_second": 15.744,
|
| 53 |
+
"eval_steps_per_second": 1.97,
|
| 54 |
+
"step": 1500
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.36845983787767134,
|
| 58 |
+
"grad_norm": 5.3738909628009424e-05,
|
| 59 |
+
"learning_rate": 4.5394252026529107e-07,
|
| 60 |
+
"loss": 0.1552,
|
| 61 |
+
"step": 2000
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.36845983787767134,
|
| 65 |
+
"eval_loss": 0.8325753808021545,
|
| 66 |
+
"eval_runtime": 76.6127,
|
| 67 |
+
"eval_samples_per_second": 15.755,
|
| 68 |
+
"eval_steps_per_second": 1.971,
|
| 69 |
+
"step": 2000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.46057479734708917,
|
| 73 |
+
"grad_norm": 404.510009765625,
|
| 74 |
+
"learning_rate": 4.4242815033161386e-07,
|
| 75 |
+
"loss": 0.1753,
|
| 76 |
+
"step": 2500
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.46057479734708917,
|
| 80 |
+
"eval_loss": 0.7550467848777771,
|
| 81 |
+
"eval_runtime": 76.8024,
|
| 82 |
+
"eval_samples_per_second": 15.716,
|
| 83 |
+
"eval_steps_per_second": 1.966,
|
| 84 |
+
"step": 2500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.552689756816507,
|
| 88 |
+
"grad_norm": 370.7579650878906,
|
| 89 |
+
"learning_rate": 4.3091378039793665e-07,
|
| 90 |
+
"loss": 0.1659,
|
| 91 |
+
"step": 3000
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.552689756816507,
|
| 95 |
+
"eval_loss": 0.719167172908783,
|
| 96 |
+
"eval_runtime": 76.6327,
|
| 97 |
+
"eval_samples_per_second": 15.75,
|
| 98 |
+
"eval_steps_per_second": 1.97,
|
| 99 |
+
"step": 3000
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.6448047162859248,
|
| 103 |
+
"grad_norm": 24.97766876220703,
|
| 104 |
+
"learning_rate": 4.193994104642594e-07,
|
| 105 |
+
"loss": 0.105,
|
| 106 |
+
"step": 3500
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.6448047162859248,
|
| 110 |
+
"eval_loss": 0.7117515802383423,
|
| 111 |
+
"eval_runtime": 77.1286,
|
| 112 |
+
"eval_samples_per_second": 15.649,
|
| 113 |
+
"eval_steps_per_second": 1.958,
|
| 114 |
+
"step": 3500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.7369196757553427,
|
| 118 |
+
"grad_norm": 305.62060546875,
|
| 119 |
+
"learning_rate": 4.0788504053058217e-07,
|
| 120 |
+
"loss": 0.1336,
|
| 121 |
+
"step": 4000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.7369196757553427,
|
| 125 |
+
"eval_loss": 0.6952972412109375,
|
| 126 |
+
"eval_runtime": 76.7253,
|
| 127 |
+
"eval_samples_per_second": 15.731,
|
| 128 |
+
"eval_steps_per_second": 1.968,
|
| 129 |
+
"step": 4000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.8290346352247605,
|
| 133 |
+
"grad_norm": 506.12237548828125,
|
| 134 |
+
"learning_rate": 3.9637067059690496e-07,
|
| 135 |
+
"loss": 0.1154,
|
| 136 |
+
"step": 4500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.8290346352247605,
|
| 140 |
+
"eval_loss": 0.6744509935379028,
|
| 141 |
+
"eval_runtime": 76.9493,
|
| 142 |
+
"eval_samples_per_second": 15.686,
|
| 143 |
+
"eval_steps_per_second": 1.962,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.9211495946941783,
|
| 148 |
+
"grad_norm": 8.823301322991028e-05,
|
| 149 |
+
"learning_rate": 3.8485630066322774e-07,
|
| 150 |
+
"loss": 0.108,
|
| 151 |
+
"step": 5000
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.9211495946941783,
|
| 155 |
+
"eval_loss": 0.6559990048408508,
|
| 156 |
+
"eval_runtime": 75.2391,
|
| 157 |
+
"eval_samples_per_second": 16.042,
|
| 158 |
+
"eval_steps_per_second": 2.007,
|
| 159 |
+
"step": 5000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.013264554163596,
|
| 163 |
+
"grad_norm": 0.0008560372516512871,
|
| 164 |
+
"learning_rate": 3.733419307295504e-07,
|
| 165 |
+
"loss": 0.1106,
|
| 166 |
+
"step": 5500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 1.013264554163596,
|
| 170 |
+
"eval_loss": 0.6367496848106384,
|
| 171 |
+
"eval_runtime": 77.0031,
|
| 172 |
+
"eval_samples_per_second": 15.675,
|
| 173 |
+
"eval_steps_per_second": 1.961,
|
| 174 |
+
"step": 5500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 1.105379513633014,
|
| 178 |
+
"grad_norm": 0.04510360211133957,
|
| 179 |
+
"learning_rate": 3.618275607958732e-07,
|
| 180 |
+
"loss": 0.0591,
|
| 181 |
+
"step": 6000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 1.105379513633014,
|
| 185 |
+
"eval_loss": 0.6259325742721558,
|
| 186 |
+
"eval_runtime": 75.2787,
|
| 187 |
+
"eval_samples_per_second": 16.034,
|
| 188 |
+
"eval_steps_per_second": 2.006,
|
| 189 |
+
"step": 6000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.1974944731024317,
|
| 193 |
+
"grad_norm": 0.0005158882704563439,
|
| 194 |
+
"learning_rate": 3.50313190862196e-07,
|
| 195 |
+
"loss": 0.0745,
|
| 196 |
+
"step": 6500
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 1.1974944731024317,
|
| 200 |
+
"eval_loss": 0.6209793090820312,
|
| 201 |
+
"eval_runtime": 77.1989,
|
| 202 |
+
"eval_samples_per_second": 15.635,
|
| 203 |
+
"eval_steps_per_second": 1.956,
|
| 204 |
+
"step": 6500
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 1.2896094325718497,
|
| 208 |
+
"grad_norm": 3.5883346072296263e-07,
|
| 209 |
+
"learning_rate": 3.387988209285188e-07,
|
| 210 |
+
"loss": 0.0502,
|
| 211 |
+
"step": 7000
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 1.2896094325718497,
|
| 215 |
+
"eval_loss": 0.6133300065994263,
|
| 216 |
+
"eval_runtime": 75.1116,
|
| 217 |
+
"eval_samples_per_second": 16.069,
|
| 218 |
+
"eval_steps_per_second": 2.01,
|
| 219 |
+
"step": 7000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.3817243920412676,
|
| 223 |
+
"grad_norm": 21.66597557067871,
|
| 224 |
+
"learning_rate": 3.272844509948415e-07,
|
| 225 |
+
"loss": 0.079,
|
| 226 |
+
"step": 7500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 1.3817243920412676,
|
| 230 |
+
"eval_loss": 0.6006803512573242,
|
| 231 |
+
"eval_runtime": 75.0086,
|
| 232 |
+
"eval_samples_per_second": 16.091,
|
| 233 |
+
"eval_steps_per_second": 2.013,
|
| 234 |
+
"step": 7500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.4738393515106853,
|
| 238 |
+
"grad_norm": 7.992870814632624e-05,
|
| 239 |
+
"learning_rate": 3.157700810611643e-07,
|
| 240 |
+
"loss": 0.0776,
|
| 241 |
+
"step": 8000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.4738393515106853,
|
| 245 |
+
"eval_loss": 0.5866427421569824,
|
| 246 |
+
"eval_runtime": 75.6702,
|
| 247 |
+
"eval_samples_per_second": 15.951,
|
| 248 |
+
"eval_steps_per_second": 1.996,
|
| 249 |
+
"step": 8000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.565954310980103,
|
| 253 |
+
"grad_norm": 0.0010381735628470778,
|
| 254 |
+
"learning_rate": 3.042557111274871e-07,
|
| 255 |
+
"loss": 0.0492,
|
| 256 |
+
"step": 8500
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.565954310980103,
|
| 260 |
+
"eval_loss": 0.5679212212562561,
|
| 261 |
+
"eval_runtime": 75.0643,
|
| 262 |
+
"eval_samples_per_second": 16.08,
|
| 263 |
+
"eval_steps_per_second": 2.012,
|
| 264 |
+
"step": 8500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.658069270449521,
|
| 268 |
+
"grad_norm": 1.733377956725235e-07,
|
| 269 |
+
"learning_rate": 2.927413411938099e-07,
|
| 270 |
+
"loss": 0.0794,
|
| 271 |
+
"step": 9000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 1.658069270449521,
|
| 275 |
+
"eval_loss": 0.5761673450469971,
|
| 276 |
+
"eval_runtime": 76.2738,
|
| 277 |
+
"eval_samples_per_second": 15.825,
|
| 278 |
+
"eval_steps_per_second": 1.98,
|
| 279 |
+
"step": 9000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.750184229918939,
|
| 283 |
+
"grad_norm": 3.1063287359023306e-11,
|
| 284 |
+
"learning_rate": 2.812269712601326e-07,
|
| 285 |
+
"loss": 0.0677,
|
| 286 |
+
"step": 9500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 1.750184229918939,
|
| 290 |
+
"eval_loss": 0.5565799474716187,
|
| 291 |
+
"eval_runtime": 75.6897,
|
| 292 |
+
"eval_samples_per_second": 15.947,
|
| 293 |
+
"eval_steps_per_second": 1.995,
|
| 294 |
+
"step": 9500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.8422991893883567,
|
| 298 |
+
"grad_norm": 0.00010032750287791714,
|
| 299 |
+
"learning_rate": 2.697126013264554e-07,
|
| 300 |
+
"loss": 0.0566,
|
| 301 |
+
"step": 10000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 1.8422991893883567,
|
| 305 |
+
"eval_loss": 0.5482199192047119,
|
| 306 |
+
"eval_runtime": 75.0029,
|
| 307 |
+
"eval_samples_per_second": 16.093,
|
| 308 |
+
"eval_steps_per_second": 2.013,
|
| 309 |
+
"step": 10000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.9344141488577744,
|
| 313 |
+
"grad_norm": 5.553330595375883e-08,
|
| 314 |
+
"learning_rate": 2.581982313927782e-07,
|
| 315 |
+
"loss": 0.0828,
|
| 316 |
+
"step": 10500
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.9344141488577744,
|
| 320 |
+
"eval_loss": 0.5500391125679016,
|
| 321 |
+
"eval_runtime": 76.2612,
|
| 322 |
+
"eval_samples_per_second": 15.827,
|
| 323 |
+
"eval_steps_per_second": 1.98,
|
| 324 |
+
"step": 10500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 2.026529108327192,
|
| 328 |
+
"grad_norm": 1.5537947319899104e-06,
|
| 329 |
+
"learning_rate": 2.466838614591009e-07,
|
| 330 |
+
"loss": 0.0573,
|
| 331 |
+
"step": 11000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 2.026529108327192,
|
| 335 |
+
"eval_loss": 0.5341849327087402,
|
| 336 |
+
"eval_runtime": 75.6131,
|
| 337 |
+
"eval_samples_per_second": 15.963,
|
| 338 |
+
"eval_steps_per_second": 1.997,
|
| 339 |
+
"step": 11000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 2.1186440677966103,
|
| 343 |
+
"grad_norm": 36.10622024536133,
|
| 344 |
+
"learning_rate": 2.3516949152542374e-07,
|
| 345 |
+
"loss": 0.0401,
|
| 346 |
+
"step": 11500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 2.1186440677966103,
|
| 350 |
+
"eval_loss": 0.5351412296295166,
|
| 351 |
+
"eval_runtime": 74.8873,
|
| 352 |
+
"eval_samples_per_second": 16.118,
|
| 353 |
+
"eval_steps_per_second": 2.016,
|
| 354 |
+
"step": 11500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.210759027266028,
|
| 358 |
+
"grad_norm": 0.49580878019332886,
|
| 359 |
+
"learning_rate": 2.236551215917465e-07,
|
| 360 |
+
"loss": 0.0152,
|
| 361 |
+
"step": 12000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.210759027266028,
|
| 365 |
+
"eval_loss": 0.5349487662315369,
|
| 366 |
+
"eval_runtime": 76.3109,
|
| 367 |
+
"eval_samples_per_second": 15.817,
|
| 368 |
+
"eval_steps_per_second": 1.979,
|
| 369 |
+
"step": 12000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 2.3028739867354457,
|
| 373 |
+
"grad_norm": 1.7469781637191772,
|
| 374 |
+
"learning_rate": 2.1214075165806928e-07,
|
| 375 |
+
"loss": 0.0638,
|
| 376 |
+
"step": 12500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 2.3028739867354457,
|
| 380 |
+
"eval_loss": 0.531768798828125,
|
| 381 |
+
"eval_runtime": 75.5712,
|
| 382 |
+
"eval_samples_per_second": 15.972,
|
| 383 |
+
"eval_steps_per_second": 1.998,
|
| 384 |
+
"step": 12500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 2.3949889462048635,
|
| 388 |
+
"grad_norm": 0.0006845183088444173,
|
| 389 |
+
"learning_rate": 2.0062638172439202e-07,
|
| 390 |
+
"loss": 0.0488,
|
| 391 |
+
"step": 13000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 2.3949889462048635,
|
| 395 |
+
"eval_loss": 0.530552089214325,
|
| 396 |
+
"eval_runtime": 75.6479,
|
| 397 |
+
"eval_samples_per_second": 15.955,
|
| 398 |
+
"eval_steps_per_second": 1.996,
|
| 399 |
+
"step": 13000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 2.4871039056742816,
|
| 403 |
+
"grad_norm": 0.0002451306500006467,
|
| 404 |
+
"learning_rate": 1.891120117907148e-07,
|
| 405 |
+
"loss": 0.0456,
|
| 406 |
+
"step": 13500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 2.4871039056742816,
|
| 410 |
+
"eval_loss": 0.5211473107337952,
|
| 411 |
+
"eval_runtime": 76.4401,
|
| 412 |
+
"eval_samples_per_second": 15.79,
|
| 413 |
+
"eval_steps_per_second": 1.975,
|
| 414 |
+
"step": 13500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 2.5792188651436994,
|
| 418 |
+
"grad_norm": 0.003609905717894435,
|
| 419 |
+
"learning_rate": 1.7759764185703757e-07,
|
| 420 |
+
"loss": 0.0264,
|
| 421 |
+
"step": 14000
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 2.5792188651436994,
|
| 425 |
+
"eval_loss": 0.5193639993667603,
|
| 426 |
+
"eval_runtime": 74.595,
|
| 427 |
+
"eval_samples_per_second": 16.181,
|
| 428 |
+
"eval_steps_per_second": 2.024,
|
| 429 |
+
"step": 14000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 2.671333824613117,
|
| 433 |
+
"grad_norm": 6.751325276610487e-09,
|
| 434 |
+
"learning_rate": 1.6608327192336035e-07,
|
| 435 |
+
"loss": 0.0381,
|
| 436 |
+
"step": 14500
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 2.671333824613117,
|
| 440 |
+
"eval_loss": 0.520633339881897,
|
| 441 |
+
"eval_runtime": 75.2937,
|
| 442 |
+
"eval_samples_per_second": 16.031,
|
| 443 |
+
"eval_steps_per_second": 2.005,
|
| 444 |
+
"step": 14500
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 2.7634487840825352,
|
| 448 |
+
"grad_norm": 8.452875022157968e-07,
|
| 449 |
+
"learning_rate": 1.5456890198968311e-07,
|
| 450 |
+
"loss": 0.0413,
|
| 451 |
+
"step": 15000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 2.7634487840825352,
|
| 455 |
+
"eval_loss": 0.5167532563209534,
|
| 456 |
+
"eval_runtime": 75.615,
|
| 457 |
+
"eval_samples_per_second": 15.962,
|
| 458 |
+
"eval_steps_per_second": 1.997,
|
| 459 |
+
"step": 15000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.855563743551953,
|
| 463 |
+
"grad_norm": 2.365756202049596e-11,
|
| 464 |
+
"learning_rate": 1.430545320560059e-07,
|
| 465 |
+
"loss": 0.0392,
|
| 466 |
+
"step": 15500
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.855563743551953,
|
| 470 |
+
"eval_loss": 0.5148531198501587,
|
| 471 |
+
"eval_runtime": 74.6217,
|
| 472 |
+
"eval_samples_per_second": 16.175,
|
| 473 |
+
"eval_steps_per_second": 2.024,
|
| 474 |
+
"step": 15500
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 2.9476787030213707,
|
| 478 |
+
"grad_norm": 0.32707780599594116,
|
| 479 |
+
"learning_rate": 1.3154016212232866e-07,
|
| 480 |
+
"loss": 0.0352,
|
| 481 |
+
"step": 16000
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 2.9476787030213707,
|
| 485 |
+
"eval_loss": 0.5112494826316833,
|
| 486 |
+
"eval_runtime": 75.142,
|
| 487 |
+
"eval_samples_per_second": 16.063,
|
| 488 |
+
"eval_steps_per_second": 2.01,
|
| 489 |
+
"step": 16000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 3.0397936624907884,
|
| 493 |
+
"grad_norm": 4.2568670810538833e-11,
|
| 494 |
+
"learning_rate": 1.2002579218865142e-07,
|
| 495 |
+
"loss": 0.0467,
|
| 496 |
+
"step": 16500
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 3.0397936624907884,
|
| 500 |
+
"eval_loss": 0.5097618699073792,
|
| 501 |
+
"eval_runtime": 75.705,
|
| 502 |
+
"eval_samples_per_second": 15.943,
|
| 503 |
+
"eval_steps_per_second": 1.995,
|
| 504 |
+
"step": 16500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 3.131908621960206,
|
| 508 |
+
"grad_norm": 7.157668679980134e-09,
|
| 509 |
+
"learning_rate": 1.085114222549742e-07,
|
| 510 |
+
"loss": 0.0366,
|
| 511 |
+
"step": 17000
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 3.131908621960206,
|
| 515 |
+
"eval_loss": 0.5088639855384827,
|
| 516 |
+
"eval_runtime": 74.5863,
|
| 517 |
+
"eval_samples_per_second": 16.183,
|
| 518 |
+
"eval_steps_per_second": 2.025,
|
| 519 |
+
"step": 17000
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 3.2240235814296243,
|
| 523 |
+
"grad_norm": 0.5768880248069763,
|
| 524 |
+
"learning_rate": 9.699705232129697e-08,
|
| 525 |
+
"loss": 0.0454,
|
| 526 |
+
"step": 17500
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 3.2240235814296243,
|
| 530 |
+
"eval_loss": 0.5103679299354553,
|
| 531 |
+
"eval_runtime": 74.9953,
|
| 532 |
+
"eval_samples_per_second": 16.094,
|
| 533 |
+
"eval_steps_per_second": 2.013,
|
| 534 |
+
"step": 17500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 3.316138540899042,
|
| 538 |
+
"grad_norm": 0.029806463047862053,
|
| 539 |
+
"learning_rate": 8.548268238761974e-08,
|
| 540 |
+
"loss": 0.0209,
|
| 541 |
+
"step": 18000
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 3.316138540899042,
|
| 545 |
+
"eval_loss": 0.5070606470108032,
|
| 546 |
+
"eval_runtime": 75.7962,
|
| 547 |
+
"eval_samples_per_second": 15.924,
|
| 548 |
+
"eval_steps_per_second": 1.992,
|
| 549 |
+
"step": 18000
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 3.4082535003684598,
|
| 553 |
+
"grad_norm": 0.09116169810295105,
|
| 554 |
+
"learning_rate": 7.396831245394252e-08,
|
| 555 |
+
"loss": 0.0636,
|
| 556 |
+
"step": 18500
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 3.4082535003684598,
|
| 560 |
+
"eval_loss": 0.5044638514518738,
|
| 561 |
+
"eval_runtime": 74.7518,
|
| 562 |
+
"eval_samples_per_second": 16.147,
|
| 563 |
+
"eval_steps_per_second": 2.02,
|
| 564 |
+
"step": 18500
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 3.500368459837878,
|
| 568 |
+
"grad_norm": 0.13625642657279968,
|
| 569 |
+
"learning_rate": 6.245394252026529e-08,
|
| 570 |
+
"loss": 0.0159,
|
| 571 |
+
"step": 19000
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 3.500368459837878,
|
| 575 |
+
"eval_loss": 0.5019211769104004,
|
| 576 |
+
"eval_runtime": 75.0004,
|
| 577 |
+
"eval_samples_per_second": 16.093,
|
| 578 |
+
"eval_steps_per_second": 2.013,
|
| 579 |
+
"step": 19000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 3.5924834193072956,
|
| 583 |
+
"grad_norm": 2.3882080313342158e-06,
|
| 584 |
+
"learning_rate": 5.0939572586588065e-08,
|
| 585 |
+
"loss": 0.0303,
|
| 586 |
+
"step": 19500
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 3.5924834193072956,
|
| 590 |
+
"eval_loss": 0.49851301312446594,
|
| 591 |
+
"eval_runtime": 75.7613,
|
| 592 |
+
"eval_samples_per_second": 15.932,
|
| 593 |
+
"eval_steps_per_second": 1.993,
|
| 594 |
+
"step": 19500
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 3.6845983787767134,
|
| 598 |
+
"grad_norm": 8.883699109674126e-09,
|
| 599 |
+
"learning_rate": 3.942520265291083e-08,
|
| 600 |
+
"loss": 0.0353,
|
| 601 |
+
"step": 20000
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 3.6845983787767134,
|
| 605 |
+
"eval_loss": 0.49747875332832336,
|
| 606 |
+
"eval_runtime": 74.5978,
|
| 607 |
+
"eval_samples_per_second": 16.18,
|
| 608 |
+
"eval_steps_per_second": 2.024,
|
| 609 |
+
"step": 20000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 3.776713338246131,
|
| 613 |
+
"grad_norm": 0.12466106563806534,
|
| 614 |
+
"learning_rate": 2.7910832719233602e-08,
|
| 615 |
+
"loss": 0.0261,
|
| 616 |
+
"step": 20500
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 3.776713338246131,
|
| 620 |
+
"eval_loss": 0.4962019622325897,
|
| 621 |
+
"eval_runtime": 75.2409,
|
| 622 |
+
"eval_samples_per_second": 16.042,
|
| 623 |
+
"eval_steps_per_second": 2.007,
|
| 624 |
+
"step": 20500
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 3.868828297715549,
|
| 628 |
+
"grad_norm": 1.0646399317693067e-07,
|
| 629 |
+
"learning_rate": 1.6396462785556373e-08,
|
| 630 |
+
"loss": 0.0291,
|
| 631 |
+
"step": 21000
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 3.868828297715549,
|
| 635 |
+
"eval_loss": 0.49561700224876404,
|
| 636 |
+
"eval_runtime": 75.8041,
|
| 637 |
+
"eval_samples_per_second": 15.923,
|
| 638 |
+
"eval_steps_per_second": 1.992,
|
| 639 |
+
"step": 21000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 3.960943257184967,
|
| 643 |
+
"grad_norm": 3.997461042737349e-11,
|
| 644 |
+
"learning_rate": 4.882092851879145e-09,
|
| 645 |
+
"loss": 0.0338,
|
| 646 |
+
"step": 21500
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 3.960943257184967,
|
| 650 |
+
"eval_loss": 0.495645672082901,
|
| 651 |
+
"eval_runtime": 74.5834,
|
| 652 |
+
"eval_samples_per_second": 16.183,
|
| 653 |
+
"eval_steps_per_second": 2.025,
|
| 654 |
+
"step": 21500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 3.6844749623178696,
|
| 658 |
+
"grad_norm": 0.022144177928566933,
|
| 659 |
+
"learning_rate": 3.944062971026629e-08,
|
| 660 |
+
"loss": 0.0432,
|
| 661 |
+
"step": 22000
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 3.6844749623178696,
|
| 665 |
+
"eval_loss": 0.17835669219493866,
|
| 666 |
+
"eval_runtime": 7.4507,
|
| 667 |
+
"eval_samples_per_second": 16.24,
|
| 668 |
+
"eval_steps_per_second": 2.147,
|
| 669 |
+
"step": 22000
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 3.7682130296432756,
|
| 673 |
+
"grad_norm": 2.4324751279891643e-07,
|
| 674 |
+
"learning_rate": 2.897337129459052e-08,
|
| 675 |
+
"loss": 0.0461,
|
| 676 |
+
"step": 22500
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 3.7682130296432756,
|
| 680 |
+
"eval_loss": 0.17669138312339783,
|
| 681 |
+
"eval_runtime": 7.5623,
|
| 682 |
+
"eval_samples_per_second": 16.0,
|
| 683 |
+
"eval_steps_per_second": 2.116,
|
| 684 |
+
"step": 22500
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 3.851951096968682,
|
| 688 |
+
"grad_norm": 4.172260048562748e-07,
|
| 689 |
+
"learning_rate": 1.8506112878914754e-08,
|
| 690 |
+
"loss": 0.0513,
|
| 691 |
+
"step": 23000
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 3.851951096968682,
|
| 695 |
+
"eval_loss": 0.1774093359708786,
|
| 696 |
+
"eval_runtime": 7.8193,
|
| 697 |
+
"eval_samples_per_second": 15.475,
|
| 698 |
+
"eval_steps_per_second": 2.046,
|
| 699 |
+
"step": 23000
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 3.935689164294088,
|
| 703 |
+
"grad_norm": 0.5274502038955688,
|
| 704 |
+
"learning_rate": 8.038854463238988e-09,
|
| 705 |
+
"loss": 0.0326,
|
| 706 |
+
"step": 23500
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 3.935689164294088,
|
| 710 |
+
"eval_loss": 0.17657655477523804,
|
| 711 |
+
"eval_runtime": 7.7889,
|
| 712 |
+
"eval_samples_per_second": 15.535,
|
| 713 |
+
"eval_steps_per_second": 2.054,
|
| 714 |
+
"step": 23500
|
| 715 |
+
}
|
| 716 |
+
],
|
| 717 |
+
"logging_steps": 500,
|
| 718 |
+
"max_steps": 23884,
|
| 719 |
+
"num_input_tokens_seen": 0,
|
| 720 |
+
"num_train_epochs": 4,
|
| 721 |
+
"save_steps": 500,
|
| 722 |
+
"stateful_callbacks": {
|
| 723 |
+
"TrainerControl": {
|
| 724 |
+
"args": {
|
| 725 |
+
"should_epoch_stop": false,
|
| 726 |
+
"should_evaluate": false,
|
| 727 |
+
"should_log": false,
|
| 728 |
+
"should_save": true,
|
| 729 |
+
"should_training_stop": false
|
| 730 |
+
},
|
| 731 |
+
"attributes": {}
|
| 732 |
+
}
|
| 733 |
+
},
|
| 734 |
+
"total_flos": 8455040771312520.0,
|
| 735 |
+
"train_batch_size": 2,
|
| 736 |
+
"trial_name": null,
|
| 737 |
+
"trial_params": null
|
| 738 |
+
}
|
checkpoint-23500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f71c1b7b2e3d63d5d414705f142c4860d9045ea3e8f1377d3c6bfb13c6fefb73
|
| 3 |
+
size 4847
|