Training in progress, step 26500, checkpoint
Browse files
checkpoint-26500/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-26500/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:04e6a2b0bbfa037c2276562c42a75e305ea5293a5eebe30cd6890d7e8d2603d1
|
| 3 |
+
size 1711848436
|
checkpoint-26500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:058adfd5737a6dd702972b5a895b30a49b53848c7252598e600d96a29bdc298a
|
| 3 |
+
size 3424043887
|
checkpoint-26500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f17cfa9c991eb6e7de239b3a28caf1e207b1e04e393abadaa128120b31341a6a
|
| 3 |
+
size 14503
|
checkpoint-26500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:38b983765e90f6c84813147bd2c640d3830fd171876b056320add189f460319a
|
| 3 |
+
size 623
|
checkpoint-26500/trainer_state.json
ADDED
|
@@ -0,0 +1,828 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8679539561271667,
|
| 3 |
+
"best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l58-l/checkpoint-26500",
|
| 4 |
+
"epoch": 2.8263651877133107,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 26500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.05332764505119454,
|
| 13 |
+
"grad_norm": 217.40109252929688,
|
| 14 |
+
"learning_rate": 4.911120591581342e-08,
|
| 15 |
+
"loss": 0.4667,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.05332764505119454,
|
| 20 |
+
"eval_loss": 1.4427576065063477,
|
| 21 |
+
"eval_runtime": 60.328,
|
| 22 |
+
"eval_samples_per_second": 16.361,
|
| 23 |
+
"eval_steps_per_second": 2.055,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.10665529010238908,
|
| 28 |
+
"grad_norm": 70.38941955566406,
|
| 29 |
+
"learning_rate": 4.822241183162685e-08,
|
| 30 |
+
"loss": 0.4534,
|
| 31 |
+
"step": 1000
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.10665529010238908,
|
| 35 |
+
"eval_loss": 1.3823094367980957,
|
| 36 |
+
"eval_runtime": 61.4884,
|
| 37 |
+
"eval_samples_per_second": 16.052,
|
| 38 |
+
"eval_steps_per_second": 2.017,
|
| 39 |
+
"step": 1000
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.1599829351535836,
|
| 43 |
+
"grad_norm": 389.9967956542969,
|
| 44 |
+
"learning_rate": 4.733361774744027e-08,
|
| 45 |
+
"loss": 0.3753,
|
| 46 |
+
"step": 1500
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.1599829351535836,
|
| 50 |
+
"eval_loss": 1.3328827619552612,
|
| 51 |
+
"eval_runtime": 63.0378,
|
| 52 |
+
"eval_samples_per_second": 15.657,
|
| 53 |
+
"eval_steps_per_second": 1.967,
|
| 54 |
+
"step": 1500
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.21331058020477817,
|
| 58 |
+
"grad_norm": 13.427445411682129,
|
| 59 |
+
"learning_rate": 4.6444823663253694e-08,
|
| 60 |
+
"loss": 0.3361,
|
| 61 |
+
"step": 2000
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.21331058020477817,
|
| 65 |
+
"eval_loss": 1.2922736406326294,
|
| 66 |
+
"eval_runtime": 63.1958,
|
| 67 |
+
"eval_samples_per_second": 15.618,
|
| 68 |
+
"eval_steps_per_second": 1.962,
|
| 69 |
+
"step": 2000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.2666382252559727,
|
| 73 |
+
"grad_norm": 320.7197265625,
|
| 74 |
+
"learning_rate": 4.555602957906712e-08,
|
| 75 |
+
"loss": 0.3597,
|
| 76 |
+
"step": 2500
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.2666382252559727,
|
| 80 |
+
"eval_loss": 1.257211685180664,
|
| 81 |
+
"eval_runtime": 63.3241,
|
| 82 |
+
"eval_samples_per_second": 15.586,
|
| 83 |
+
"eval_steps_per_second": 1.958,
|
| 84 |
+
"step": 2500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.3199658703071672,
|
| 88 |
+
"grad_norm": 0.0019184721168130636,
|
| 89 |
+
"learning_rate": 4.4667235494880546e-08,
|
| 90 |
+
"loss": 0.3042,
|
| 91 |
+
"step": 3000
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.3199658703071672,
|
| 95 |
+
"eval_loss": 1.2240594625473022,
|
| 96 |
+
"eval_runtime": 63.5118,
|
| 97 |
+
"eval_samples_per_second": 15.54,
|
| 98 |
+
"eval_steps_per_second": 1.952,
|
| 99 |
+
"step": 3000
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.37329351535836175,
|
| 103 |
+
"grad_norm": 11.738512992858887,
|
| 104 |
+
"learning_rate": 4.3778441410693965e-08,
|
| 105 |
+
"loss": 0.3252,
|
| 106 |
+
"step": 3500
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.37329351535836175,
|
| 110 |
+
"eval_loss": 1.1988952159881592,
|
| 111 |
+
"eval_runtime": 63.5384,
|
| 112 |
+
"eval_samples_per_second": 15.534,
|
| 113 |
+
"eval_steps_per_second": 1.952,
|
| 114 |
+
"step": 3500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.42662116040955633,
|
| 118 |
+
"grad_norm": 5.839956283569336,
|
| 119 |
+
"learning_rate": 4.28896473265074e-08,
|
| 120 |
+
"loss": 0.2439,
|
| 121 |
+
"step": 4000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.42662116040955633,
|
| 125 |
+
"eval_loss": 1.1730674505233765,
|
| 126 |
+
"eval_runtime": 62.3497,
|
| 127 |
+
"eval_samples_per_second": 15.83,
|
| 128 |
+
"eval_steps_per_second": 1.989,
|
| 129 |
+
"step": 4000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.47994880546075086,
|
| 133 |
+
"grad_norm": 154.76502990722656,
|
| 134 |
+
"learning_rate": 4.2000853242320816e-08,
|
| 135 |
+
"loss": 0.3004,
|
| 136 |
+
"step": 4500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.47994880546075086,
|
| 140 |
+
"eval_loss": 1.1497682332992554,
|
| 141 |
+
"eval_runtime": 63.5491,
|
| 142 |
+
"eval_samples_per_second": 15.531,
|
| 143 |
+
"eval_steps_per_second": 1.951,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.5332764505119454,
|
| 148 |
+
"grad_norm": 0.13150489330291748,
|
| 149 |
+
"learning_rate": 4.111205915813424e-08,
|
| 150 |
+
"loss": 0.3205,
|
| 151 |
+
"step": 5000
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.5332764505119454,
|
| 155 |
+
"eval_loss": 1.1258739233016968,
|
| 156 |
+
"eval_runtime": 62.1884,
|
| 157 |
+
"eval_samples_per_second": 15.871,
|
| 158 |
+
"eval_steps_per_second": 1.994,
|
| 159 |
+
"step": 5000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.58660409556314,
|
| 163 |
+
"grad_norm": 0.3719326853752136,
|
| 164 |
+
"learning_rate": 4.022326507394767e-08,
|
| 165 |
+
"loss": 0.3091,
|
| 166 |
+
"step": 5500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.58660409556314,
|
| 170 |
+
"eval_loss": 1.1053005456924438,
|
| 171 |
+
"eval_runtime": 63.437,
|
| 172 |
+
"eval_samples_per_second": 15.559,
|
| 173 |
+
"eval_steps_per_second": 1.955,
|
| 174 |
+
"step": 5500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.6399317406143344,
|
| 178 |
+
"grad_norm": 170.89846801757812,
|
| 179 |
+
"learning_rate": 3.9334470989761093e-08,
|
| 180 |
+
"loss": 0.2657,
|
| 181 |
+
"step": 6000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.6399317406143344,
|
| 185 |
+
"eval_loss": 1.0857468843460083,
|
| 186 |
+
"eval_runtime": 62.4055,
|
| 187 |
+
"eval_samples_per_second": 15.816,
|
| 188 |
+
"eval_steps_per_second": 1.987,
|
| 189 |
+
"step": 6000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.693259385665529,
|
| 193 |
+
"grad_norm": 323.0997619628906,
|
| 194 |
+
"learning_rate": 3.844567690557451e-08,
|
| 195 |
+
"loss": 0.2251,
|
| 196 |
+
"step": 6500
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.693259385665529,
|
| 200 |
+
"eval_loss": 1.0695208311080933,
|
| 201 |
+
"eval_runtime": 63.1085,
|
| 202 |
+
"eval_samples_per_second": 15.64,
|
| 203 |
+
"eval_steps_per_second": 1.965,
|
| 204 |
+
"step": 6500
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.7465870307167235,
|
| 208 |
+
"grad_norm": 0.13254186511039734,
|
| 209 |
+
"learning_rate": 3.7556882821387945e-08,
|
| 210 |
+
"loss": 0.248,
|
| 211 |
+
"step": 7000
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.7465870307167235,
|
| 215 |
+
"eval_loss": 1.0535719394683838,
|
| 216 |
+
"eval_runtime": 62.3504,
|
| 217 |
+
"eval_samples_per_second": 15.83,
|
| 218 |
+
"eval_steps_per_second": 1.989,
|
| 219 |
+
"step": 7000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.7999146757679181,
|
| 223 |
+
"grad_norm": 63.14700698852539,
|
| 224 |
+
"learning_rate": 3.6668088737201364e-08,
|
| 225 |
+
"loss": 0.2745,
|
| 226 |
+
"step": 7500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.7999146757679181,
|
| 230 |
+
"eval_loss": 1.0388950109481812,
|
| 231 |
+
"eval_runtime": 62.2311,
|
| 232 |
+
"eval_samples_per_second": 15.86,
|
| 233 |
+
"eval_steps_per_second": 1.993,
|
| 234 |
+
"step": 7500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.8532423208191127,
|
| 238 |
+
"grad_norm": 623.6299438476562,
|
| 239 |
+
"learning_rate": 3.577929465301478e-08,
|
| 240 |
+
"loss": 0.2857,
|
| 241 |
+
"step": 8000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.8532423208191127,
|
| 245 |
+
"eval_loss": 1.0260424613952637,
|
| 246 |
+
"eval_runtime": 62.2792,
|
| 247 |
+
"eval_samples_per_second": 15.848,
|
| 248 |
+
"eval_steps_per_second": 1.991,
|
| 249 |
+
"step": 8000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.9065699658703071,
|
| 253 |
+
"grad_norm": 155.54579162597656,
|
| 254 |
+
"learning_rate": 3.4890500568828216e-08,
|
| 255 |
+
"loss": 0.2365,
|
| 256 |
+
"step": 8500
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.9065699658703071,
|
| 260 |
+
"eval_loss": 1.0136038064956665,
|
| 261 |
+
"eval_runtime": 62.583,
|
| 262 |
+
"eval_samples_per_second": 15.771,
|
| 263 |
+
"eval_steps_per_second": 1.981,
|
| 264 |
+
"step": 8500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.9598976109215017,
|
| 268 |
+
"grad_norm": 76.14079284667969,
|
| 269 |
+
"learning_rate": 3.4001706484641635e-08,
|
| 270 |
+
"loss": 0.2382,
|
| 271 |
+
"step": 9000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.9598976109215017,
|
| 275 |
+
"eval_loss": 1.0034172534942627,
|
| 276 |
+
"eval_runtime": 63.5592,
|
| 277 |
+
"eval_samples_per_second": 15.529,
|
| 278 |
+
"eval_steps_per_second": 1.951,
|
| 279 |
+
"step": 9000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.0132252559726962,
|
| 283 |
+
"grad_norm": 571.6434936523438,
|
| 284 |
+
"learning_rate": 3.311291240045506e-08,
|
| 285 |
+
"loss": 0.2419,
|
| 286 |
+
"step": 9500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 1.0132252559726962,
|
| 290 |
+
"eval_loss": 0.994463324546814,
|
| 291 |
+
"eval_runtime": 61.8407,
|
| 292 |
+
"eval_samples_per_second": 15.96,
|
| 293 |
+
"eval_steps_per_second": 2.005,
|
| 294 |
+
"step": 9500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.0665529010238908,
|
| 298 |
+
"grad_norm": 540.1460571289062,
|
| 299 |
+
"learning_rate": 3.2224118316268486e-08,
|
| 300 |
+
"loss": 0.2047,
|
| 301 |
+
"step": 10000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 1.0665529010238908,
|
| 305 |
+
"eval_loss": 0.9851205348968506,
|
| 306 |
+
"eval_runtime": 62.1686,
|
| 307 |
+
"eval_samples_per_second": 15.876,
|
| 308 |
+
"eval_steps_per_second": 1.995,
|
| 309 |
+
"step": 10000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.1198805460750854,
|
| 313 |
+
"grad_norm": 6.2389655113220215,
|
| 314 |
+
"learning_rate": 3.133532423208191e-08,
|
| 315 |
+
"loss": 0.2133,
|
| 316 |
+
"step": 10500
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.1198805460750854,
|
| 320 |
+
"eval_loss": 0.9759647250175476,
|
| 321 |
+
"eval_runtime": 62.8691,
|
| 322 |
+
"eval_samples_per_second": 15.699,
|
| 323 |
+
"eval_steps_per_second": 1.972,
|
| 324 |
+
"step": 10500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.17320819112628,
|
| 328 |
+
"grad_norm": 321.5069885253906,
|
| 329 |
+
"learning_rate": 3.044653014789533e-08,
|
| 330 |
+
"loss": 0.207,
|
| 331 |
+
"step": 11000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.17320819112628,
|
| 335 |
+
"eval_loss": 0.9683472514152527,
|
| 336 |
+
"eval_runtime": 61.915,
|
| 337 |
+
"eval_samples_per_second": 15.941,
|
| 338 |
+
"eval_steps_per_second": 2.003,
|
| 339 |
+
"step": 11000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.2265358361774745,
|
| 343 |
+
"grad_norm": 0.0028632767498493195,
|
| 344 |
+
"learning_rate": 2.955773606370876e-08,
|
| 345 |
+
"loss": 0.2369,
|
| 346 |
+
"step": 11500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.2265358361774745,
|
| 350 |
+
"eval_loss": 0.9597958922386169,
|
| 351 |
+
"eval_runtime": 62.4216,
|
| 352 |
+
"eval_samples_per_second": 15.812,
|
| 353 |
+
"eval_steps_per_second": 1.986,
|
| 354 |
+
"step": 11500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.2798634812286689,
|
| 358 |
+
"grad_norm": 1.046075701713562,
|
| 359 |
+
"learning_rate": 2.8668941979522183e-08,
|
| 360 |
+
"loss": 0.1834,
|
| 361 |
+
"step": 12000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 1.2798634812286689,
|
| 365 |
+
"eval_loss": 0.9528058767318726,
|
| 366 |
+
"eval_runtime": 62.9687,
|
| 367 |
+
"eval_samples_per_second": 15.674,
|
| 368 |
+
"eval_steps_per_second": 1.969,
|
| 369 |
+
"step": 12000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 1.3331911262798635,
|
| 373 |
+
"grad_norm": 0.01093329768627882,
|
| 374 |
+
"learning_rate": 2.778014789533561e-08,
|
| 375 |
+
"loss": 0.2002,
|
| 376 |
+
"step": 12500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 1.3331911262798635,
|
| 380 |
+
"eval_loss": 0.9465218782424927,
|
| 381 |
+
"eval_runtime": 62.0222,
|
| 382 |
+
"eval_samples_per_second": 15.914,
|
| 383 |
+
"eval_steps_per_second": 1.999,
|
| 384 |
+
"step": 12500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.386518771331058,
|
| 388 |
+
"grad_norm": 0.13472874462604523,
|
| 389 |
+
"learning_rate": 2.689135381114903e-08,
|
| 390 |
+
"loss": 0.1918,
|
| 391 |
+
"step": 13000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 1.386518771331058,
|
| 395 |
+
"eval_loss": 0.9417626857757568,
|
| 396 |
+
"eval_runtime": 62.4603,
|
| 397 |
+
"eval_samples_per_second": 15.802,
|
| 398 |
+
"eval_steps_per_second": 1.985,
|
| 399 |
+
"step": 13000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 1.4398464163822526,
|
| 403 |
+
"grad_norm": 19.54829216003418,
|
| 404 |
+
"learning_rate": 2.600255972696246e-08,
|
| 405 |
+
"loss": 0.2264,
|
| 406 |
+
"step": 13500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 1.4398464163822526,
|
| 410 |
+
"eval_loss": 0.9353039860725403,
|
| 411 |
+
"eval_runtime": 63.2324,
|
| 412 |
+
"eval_samples_per_second": 15.609,
|
| 413 |
+
"eval_steps_per_second": 1.961,
|
| 414 |
+
"step": 13500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 1.493174061433447,
|
| 418 |
+
"grad_norm": 6.569502353668213,
|
| 419 |
+
"learning_rate": 2.511376564277588e-08,
|
| 420 |
+
"loss": 0.171,
|
| 421 |
+
"step": 14000
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 1.493174061433447,
|
| 425 |
+
"eval_loss": 0.930813729763031,
|
| 426 |
+
"eval_runtime": 61.9703,
|
| 427 |
+
"eval_samples_per_second": 15.927,
|
| 428 |
+
"eval_steps_per_second": 2.001,
|
| 429 |
+
"step": 14000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.5465017064846416,
|
| 433 |
+
"grad_norm": 153.30908203125,
|
| 434 |
+
"learning_rate": 2.4224971558589305e-08,
|
| 435 |
+
"loss": 0.2029,
|
| 436 |
+
"step": 14500
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.5465017064846416,
|
| 440 |
+
"eval_loss": 0.9262459874153137,
|
| 441 |
+
"eval_runtime": 62.3851,
|
| 442 |
+
"eval_samples_per_second": 15.821,
|
| 443 |
+
"eval_steps_per_second": 1.988,
|
| 444 |
+
"step": 14500
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.5998293515358362,
|
| 448 |
+
"grad_norm": 578.005615234375,
|
| 449 |
+
"learning_rate": 2.333617747440273e-08,
|
| 450 |
+
"loss": 0.173,
|
| 451 |
+
"step": 15000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.5998293515358362,
|
| 455 |
+
"eval_loss": 0.9224158525466919,
|
| 456 |
+
"eval_runtime": 63.0306,
|
| 457 |
+
"eval_samples_per_second": 15.659,
|
| 458 |
+
"eval_steps_per_second": 1.967,
|
| 459 |
+
"step": 15000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 1.6531569965870307,
|
| 463 |
+
"grad_norm": 38.18622970581055,
|
| 464 |
+
"learning_rate": 2.2447383390216153e-08,
|
| 465 |
+
"loss": 0.2126,
|
| 466 |
+
"step": 15500
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 1.6531569965870307,
|
| 470 |
+
"eval_loss": 0.9170305132865906,
|
| 471 |
+
"eval_runtime": 61.9661,
|
| 472 |
+
"eval_samples_per_second": 15.928,
|
| 473 |
+
"eval_steps_per_second": 2.001,
|
| 474 |
+
"step": 15500
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 1.7064846416382253,
|
| 478 |
+
"grad_norm": 279.6297302246094,
|
| 479 |
+
"learning_rate": 2.155858930602958e-08,
|
| 480 |
+
"loss": 0.1933,
|
| 481 |
+
"step": 16000
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 1.7064846416382253,
|
| 485 |
+
"eval_loss": 0.9142164587974548,
|
| 486 |
+
"eval_runtime": 62.0524,
|
| 487 |
+
"eval_samples_per_second": 15.906,
|
| 488 |
+
"eval_steps_per_second": 1.998,
|
| 489 |
+
"step": 16000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 1.75981228668942,
|
| 493 |
+
"grad_norm": 32.153900146484375,
|
| 494 |
+
"learning_rate": 2.0669795221843004e-08,
|
| 495 |
+
"loss": 0.1911,
|
| 496 |
+
"step": 16500
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 1.75981228668942,
|
| 500 |
+
"eval_loss": 0.9099463820457458,
|
| 501 |
+
"eval_runtime": 63.0816,
|
| 502 |
+
"eval_samples_per_second": 15.646,
|
| 503 |
+
"eval_steps_per_second": 1.966,
|
| 504 |
+
"step": 16500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.8131399317406145,
|
| 508 |
+
"grad_norm": 1.433034062385559,
|
| 509 |
+
"learning_rate": 1.9781001137656427e-08,
|
| 510 |
+
"loss": 0.1809,
|
| 511 |
+
"step": 17000
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 1.8131399317406145,
|
| 515 |
+
"eval_loss": 0.9075033664703369,
|
| 516 |
+
"eval_runtime": 62.1223,
|
| 517 |
+
"eval_samples_per_second": 15.888,
|
| 518 |
+
"eval_steps_per_second": 1.996,
|
| 519 |
+
"step": 17000
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 1.8664675767918089,
|
| 523 |
+
"grad_norm": 211.28334045410156,
|
| 524 |
+
"learning_rate": 1.8892207053469853e-08,
|
| 525 |
+
"loss": 0.2699,
|
| 526 |
+
"step": 17500
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 1.8664675767918089,
|
| 530 |
+
"eval_loss": 0.9023045897483826,
|
| 531 |
+
"eval_runtime": 62.2896,
|
| 532 |
+
"eval_samples_per_second": 15.845,
|
| 533 |
+
"eval_steps_per_second": 1.991,
|
| 534 |
+
"step": 17500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.9197952218430034,
|
| 538 |
+
"grad_norm": 0.22130072116851807,
|
| 539 |
+
"learning_rate": 1.800341296928328e-08,
|
| 540 |
+
"loss": 0.1682,
|
| 541 |
+
"step": 18000
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.9197952218430034,
|
| 545 |
+
"eval_loss": 0.8992673754692078,
|
| 546 |
+
"eval_runtime": 63.0217,
|
| 547 |
+
"eval_samples_per_second": 15.661,
|
| 548 |
+
"eval_steps_per_second": 1.968,
|
| 549 |
+
"step": 18000
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.9731228668941978,
|
| 553 |
+
"grad_norm": 463.95916748046875,
|
| 554 |
+
"learning_rate": 1.7114618885096697e-08,
|
| 555 |
+
"loss": 0.2248,
|
| 556 |
+
"step": 18500
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 1.9731228668941978,
|
| 560 |
+
"eval_loss": 0.8949461579322815,
|
| 561 |
+
"eval_runtime": 62.3865,
|
| 562 |
+
"eval_samples_per_second": 15.821,
|
| 563 |
+
"eval_steps_per_second": 1.988,
|
| 564 |
+
"step": 18500
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.0264505119453924,
|
| 568 |
+
"grad_norm": 257.9815979003906,
|
| 569 |
+
"learning_rate": 1.6225824800910123e-08,
|
| 570 |
+
"loss": 0.1892,
|
| 571 |
+
"step": 19000
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.0264505119453924,
|
| 575 |
+
"eval_loss": 0.892412543296814,
|
| 576 |
+
"eval_runtime": 62.3214,
|
| 577 |
+
"eval_samples_per_second": 15.837,
|
| 578 |
+
"eval_steps_per_second": 1.99,
|
| 579 |
+
"step": 19000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 2.079778156996587,
|
| 583 |
+
"grad_norm": 50.01844787597656,
|
| 584 |
+
"learning_rate": 1.533703071672355e-08,
|
| 585 |
+
"loss": 0.1907,
|
| 586 |
+
"step": 19500
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 2.079778156996587,
|
| 590 |
+
"eval_loss": 0.8893100619316101,
|
| 591 |
+
"eval_runtime": 62.9983,
|
| 592 |
+
"eval_samples_per_second": 15.667,
|
| 593 |
+
"eval_steps_per_second": 1.968,
|
| 594 |
+
"step": 19500
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 2.1331058020477816,
|
| 598 |
+
"grad_norm": 1.0831713676452637,
|
| 599 |
+
"learning_rate": 1.4448236632536973e-08,
|
| 600 |
+
"loss": 0.1661,
|
| 601 |
+
"step": 20000
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 2.1331058020477816,
|
| 605 |
+
"eval_loss": 0.887154757976532,
|
| 606 |
+
"eval_runtime": 62.2417,
|
| 607 |
+
"eval_samples_per_second": 15.858,
|
| 608 |
+
"eval_steps_per_second": 1.992,
|
| 609 |
+
"step": 20000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 2.186433447098976,
|
| 613 |
+
"grad_norm": 0.08456582576036453,
|
| 614 |
+
"learning_rate": 1.3559442548350397e-08,
|
| 615 |
+
"loss": 0.1852,
|
| 616 |
+
"step": 20500
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 2.186433447098976,
|
| 620 |
+
"eval_loss": 0.8855507969856262,
|
| 621 |
+
"eval_runtime": 61.7252,
|
| 622 |
+
"eval_samples_per_second": 15.99,
|
| 623 |
+
"eval_steps_per_second": 2.009,
|
| 624 |
+
"step": 20500
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 2.2397610921501707,
|
| 628 |
+
"grad_norm": 19.829164505004883,
|
| 629 |
+
"learning_rate": 1.2670648464163821e-08,
|
| 630 |
+
"loss": 0.1561,
|
| 631 |
+
"step": 21000
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 2.2397610921501707,
|
| 635 |
+
"eval_loss": 0.8829911947250366,
|
| 636 |
+
"eval_runtime": 62.7443,
|
| 637 |
+
"eval_samples_per_second": 15.731,
|
| 638 |
+
"eval_steps_per_second": 1.976,
|
| 639 |
+
"step": 21000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 2.2930887372013653,
|
| 643 |
+
"grad_norm": 0.12170098721981049,
|
| 644 |
+
"learning_rate": 1.1781854379977247e-08,
|
| 645 |
+
"loss": 0.2146,
|
| 646 |
+
"step": 21500
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 2.2930887372013653,
|
| 650 |
+
"eval_loss": 0.8807759881019592,
|
| 651 |
+
"eval_runtime": 62.1264,
|
| 652 |
+
"eval_samples_per_second": 15.887,
|
| 653 |
+
"eval_steps_per_second": 1.996,
|
| 654 |
+
"step": 21500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 2.34641638225256,
|
| 658 |
+
"grad_norm": 0.0019679605029523373,
|
| 659 |
+
"learning_rate": 1.0893060295790671e-08,
|
| 660 |
+
"loss": 0.16,
|
| 661 |
+
"step": 22000
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 2.34641638225256,
|
| 665 |
+
"eval_loss": 0.8785866498947144,
|
| 666 |
+
"eval_runtime": 62.1778,
|
| 667 |
+
"eval_samples_per_second": 15.874,
|
| 668 |
+
"eval_steps_per_second": 1.994,
|
| 669 |
+
"step": 22000
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 2.399744027303754,
|
| 673 |
+
"grad_norm": 0.0032679687719792128,
|
| 674 |
+
"learning_rate": 1.0004266211604095e-08,
|
| 675 |
+
"loss": 0.1904,
|
| 676 |
+
"step": 22500
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 2.399744027303754,
|
| 680 |
+
"eval_loss": 0.8765972852706909,
|
| 681 |
+
"eval_runtime": 63.05,
|
| 682 |
+
"eval_samples_per_second": 15.654,
|
| 683 |
+
"eval_steps_per_second": 1.967,
|
| 684 |
+
"step": 22500
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 2.453071672354949,
|
| 688 |
+
"grad_norm": 0.3566090762615204,
|
| 689 |
+
"learning_rate": 9.11547212741752e-09,
|
| 690 |
+
"loss": 0.1746,
|
| 691 |
+
"step": 23000
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 2.453071672354949,
|
| 695 |
+
"eval_loss": 0.874836266040802,
|
| 696 |
+
"eval_runtime": 62.6532,
|
| 697 |
+
"eval_samples_per_second": 15.753,
|
| 698 |
+
"eval_steps_per_second": 1.979,
|
| 699 |
+
"step": 23000
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 2.506399317406143,
|
| 703 |
+
"grad_norm": 14.910808563232422,
|
| 704 |
+
"learning_rate": 8.226678043230943e-09,
|
| 705 |
+
"loss": 0.1767,
|
| 706 |
+
"step": 23500
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 2.506399317406143,
|
| 710 |
+
"eval_loss": 0.8737635612487793,
|
| 711 |
+
"eval_runtime": 62.3335,
|
| 712 |
+
"eval_samples_per_second": 15.834,
|
| 713 |
+
"eval_steps_per_second": 1.989,
|
| 714 |
+
"step": 23500
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 2.5597269624573378,
|
| 718 |
+
"grad_norm": 199.9237518310547,
|
| 719 |
+
"learning_rate": 7.3378839590443675e-09,
|
| 720 |
+
"loss": 0.1922,
|
| 721 |
+
"step": 24000
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 2.5597269624573378,
|
| 725 |
+
"eval_loss": 0.8720763921737671,
|
| 726 |
+
"eval_runtime": 62.5247,
|
| 727 |
+
"eval_samples_per_second": 15.786,
|
| 728 |
+
"eval_steps_per_second": 1.983,
|
| 729 |
+
"step": 24000
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 2.6130546075085324,
|
| 733 |
+
"grad_norm": 3.2796316146850586,
|
| 734 |
+
"learning_rate": 6.4490898748577924e-09,
|
| 735 |
+
"loss": 0.1556,
|
| 736 |
+
"step": 24500
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 2.6130546075085324,
|
| 740 |
+
"eval_loss": 0.8707790970802307,
|
| 741 |
+
"eval_runtime": 61.7705,
|
| 742 |
+
"eval_samples_per_second": 15.978,
|
| 743 |
+
"eval_steps_per_second": 2.007,
|
| 744 |
+
"step": 24500
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 2.666382252559727,
|
| 748 |
+
"grad_norm": 124.43791961669922,
|
| 749 |
+
"learning_rate": 5.560295790671217e-09,
|
| 750 |
+
"loss": 0.1388,
|
| 751 |
+
"step": 25000
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 2.666382252559727,
|
| 755 |
+
"eval_loss": 0.8700253963470459,
|
| 756 |
+
"eval_runtime": 62.0486,
|
| 757 |
+
"eval_samples_per_second": 15.907,
|
| 758 |
+
"eval_steps_per_second": 1.998,
|
| 759 |
+
"step": 25000
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 2.7197098976109215,
|
| 763 |
+
"grad_norm": 0.05777175351977348,
|
| 764 |
+
"learning_rate": 4.6715017064846414e-09,
|
| 765 |
+
"loss": 0.1766,
|
| 766 |
+
"step": 25500
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 2.7197098976109215,
|
| 770 |
+
"eval_loss": 0.8693736791610718,
|
| 771 |
+
"eval_runtime": 63.5834,
|
| 772 |
+
"eval_samples_per_second": 15.523,
|
| 773 |
+
"eval_steps_per_second": 1.95,
|
| 774 |
+
"step": 25500
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 2.773037542662116,
|
| 778 |
+
"grad_norm": 110.81836700439453,
|
| 779 |
+
"learning_rate": 3.782707622298066e-09,
|
| 780 |
+
"loss": 0.1615,
|
| 781 |
+
"step": 26000
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 2.773037542662116,
|
| 785 |
+
"eval_loss": 0.8683955669403076,
|
| 786 |
+
"eval_runtime": 62.1251,
|
| 787 |
+
"eval_samples_per_second": 15.887,
|
| 788 |
+
"eval_steps_per_second": 1.996,
|
| 789 |
+
"step": 26000
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 2.8263651877133107,
|
| 793 |
+
"grad_norm": 3.5993499755859375,
|
| 794 |
+
"learning_rate": 2.8939135381114905e-09,
|
| 795 |
+
"loss": 0.1591,
|
| 796 |
+
"step": 26500
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 2.8263651877133107,
|
| 800 |
+
"eval_loss": 0.8679539561271667,
|
| 801 |
+
"eval_runtime": 62.0886,
|
| 802 |
+
"eval_samples_per_second": 15.897,
|
| 803 |
+
"eval_steps_per_second": 1.997,
|
| 804 |
+
"step": 26500
|
| 805 |
+
}
|
| 806 |
+
],
|
| 807 |
+
"logging_steps": 500,
|
| 808 |
+
"max_steps": 28128,
|
| 809 |
+
"num_input_tokens_seen": 0,
|
| 810 |
+
"num_train_epochs": 3,
|
| 811 |
+
"save_steps": 500,
|
| 812 |
+
"stateful_callbacks": {
|
| 813 |
+
"TrainerControl": {
|
| 814 |
+
"args": {
|
| 815 |
+
"should_epoch_stop": false,
|
| 816 |
+
"should_evaluate": false,
|
| 817 |
+
"should_log": false,
|
| 818 |
+
"should_save": true,
|
| 819 |
+
"should_training_stop": false
|
| 820 |
+
},
|
| 821 |
+
"attributes": {}
|
| 822 |
+
}
|
| 823 |
+
},
|
| 824 |
+
"total_flos": 9534315825092520.0,
|
| 825 |
+
"train_batch_size": 2,
|
| 826 |
+
"trial_name": null,
|
| 827 |
+
"trial_params": null
|
| 828 |
+
}
|
checkpoint-26500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:590a3eed3f4b2ba9fcb0b831d5e4c3f2410a46be8f57155ebf9fdd97c6c96524
|
| 3 |
+
size 4847
|