Training in progress, step 28500, checkpoint
Browse files
checkpoint-28500/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-28500/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f9c68c9821135d68a882fe4ca89e0a2363377877916e00f48c2a6db19196f774
|
| 3 |
+
size 1711848436
|
checkpoint-28500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1dab2537fd12ae9ce2f688dcddf242c535d0f30d878eaad17ff3438a2f2ec1b2
|
| 3 |
+
size 3424043887
|
checkpoint-28500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bf2f313cc8e80c2a2ca71e15c033fb1b26de83782da32da7d5692888ea2cf69d
|
| 3 |
+
size 14503
|
checkpoint-28500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e1b405a10752b6669ead8997723381093a5006c141b74f8c2b19b3aea7f56e3e
|
| 3 |
+
size 623
|
checkpoint-28500/trainer_state.json
ADDED
|
@@ -0,0 +1,888 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8165684938430786,
|
| 3 |
+
"best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e4l58-l/checkpoint-28500",
|
| 4 |
+
"epoch": 3.0396757679180886,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 28500,
|
| 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": 217.55360412597656,
|
| 14 |
+
"learning_rate": 4.933340443686007e-08,
|
| 15 |
+
"loss": 0.4667,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.05332764505119454,
|
| 20 |
+
"eval_loss": 1.4426143169403076,
|
| 21 |
+
"eval_runtime": 61.9901,
|
| 22 |
+
"eval_samples_per_second": 15.922,
|
| 23 |
+
"eval_steps_per_second": 2.0,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.10665529010238908,
|
| 28 |
+
"grad_norm": 73.73340606689453,
|
| 29 |
+
"learning_rate": 4.8666808873720136e-08,
|
| 30 |
+
"loss": 0.4532,
|
| 31 |
+
"step": 1000
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.10665529010238908,
|
| 35 |
+
"eval_loss": 1.3815597295761108,
|
| 36 |
+
"eval_runtime": 62.4624,
|
| 37 |
+
"eval_samples_per_second": 15.801,
|
| 38 |
+
"eval_steps_per_second": 1.985,
|
| 39 |
+
"step": 1000
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.1599829351535836,
|
| 43 |
+
"grad_norm": 471.4011535644531,
|
| 44 |
+
"learning_rate": 4.80002133105802e-08,
|
| 45 |
+
"loss": 0.3749,
|
| 46 |
+
"step": 1500
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.1599829351535836,
|
| 50 |
+
"eval_loss": 1.3310539722442627,
|
| 51 |
+
"eval_runtime": 63.3918,
|
| 52 |
+
"eval_samples_per_second": 15.57,
|
| 53 |
+
"eval_steps_per_second": 1.956,
|
| 54 |
+
"step": 1500
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.21331058020477817,
|
| 58 |
+
"grad_norm": 11.962693214416504,
|
| 59 |
+
"learning_rate": 4.733361774744027e-08,
|
| 60 |
+
"loss": 0.336,
|
| 61 |
+
"step": 2000
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.21331058020477817,
|
| 65 |
+
"eval_loss": 1.2890639305114746,
|
| 66 |
+
"eval_runtime": 63.177,
|
| 67 |
+
"eval_samples_per_second": 15.623,
|
| 68 |
+
"eval_steps_per_second": 1.963,
|
| 69 |
+
"step": 2000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.2666382252559727,
|
| 73 |
+
"grad_norm": 375.1609191894531,
|
| 74 |
+
"learning_rate": 4.666702218430034e-08,
|
| 75 |
+
"loss": 0.3585,
|
| 76 |
+
"step": 2500
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.2666382252559727,
|
| 80 |
+
"eval_loss": 1.2536433935165405,
|
| 81 |
+
"eval_runtime": 63.1512,
|
| 82 |
+
"eval_samples_per_second": 15.629,
|
| 83 |
+
"eval_steps_per_second": 1.964,
|
| 84 |
+
"step": 2500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.3199658703071672,
|
| 88 |
+
"grad_norm": 0.0022936267778277397,
|
| 89 |
+
"learning_rate": 4.600042662116041e-08,
|
| 90 |
+
"loss": 0.303,
|
| 91 |
+
"step": 3000
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.3199658703071672,
|
| 95 |
+
"eval_loss": 1.2202869653701782,
|
| 96 |
+
"eval_runtime": 63.5857,
|
| 97 |
+
"eval_samples_per_second": 15.522,
|
| 98 |
+
"eval_steps_per_second": 1.95,
|
| 99 |
+
"step": 3000
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.37329351535836175,
|
| 103 |
+
"grad_norm": 15.805087089538574,
|
| 104 |
+
"learning_rate": 4.5333831058020476e-08,
|
| 105 |
+
"loss": 0.3242,
|
| 106 |
+
"step": 3500
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.37329351535836175,
|
| 110 |
+
"eval_loss": 1.195627212524414,
|
| 111 |
+
"eval_runtime": 63.4024,
|
| 112 |
+
"eval_samples_per_second": 15.567,
|
| 113 |
+
"eval_steps_per_second": 1.956,
|
| 114 |
+
"step": 3500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.42662116040955633,
|
| 118 |
+
"grad_norm": 5.377908229827881,
|
| 119 |
+
"learning_rate": 4.4667235494880546e-08,
|
| 120 |
+
"loss": 0.2427,
|
| 121 |
+
"step": 4000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.42662116040955633,
|
| 125 |
+
"eval_loss": 1.169384241104126,
|
| 126 |
+
"eval_runtime": 63.5082,
|
| 127 |
+
"eval_samples_per_second": 15.541,
|
| 128 |
+
"eval_steps_per_second": 1.953,
|
| 129 |
+
"step": 4000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.47994880546075086,
|
| 133 |
+
"grad_norm": 150.5334930419922,
|
| 134 |
+
"learning_rate": 4.4000639931740615e-08,
|
| 135 |
+
"loss": 0.2993,
|
| 136 |
+
"step": 4500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.47994880546075086,
|
| 140 |
+
"eval_loss": 1.145558476448059,
|
| 141 |
+
"eval_runtime": 62.3888,
|
| 142 |
+
"eval_samples_per_second": 15.82,
|
| 143 |
+
"eval_steps_per_second": 1.988,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.5332764505119454,
|
| 148 |
+
"grad_norm": 0.12246542423963547,
|
| 149 |
+
"learning_rate": 4.333404436860068e-08,
|
| 150 |
+
"loss": 0.3183,
|
| 151 |
+
"step": 5000
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.5332764505119454,
|
| 155 |
+
"eval_loss": 1.1201218366622925,
|
| 156 |
+
"eval_runtime": 63.7154,
|
| 157 |
+
"eval_samples_per_second": 15.491,
|
| 158 |
+
"eval_steps_per_second": 1.946,
|
| 159 |
+
"step": 5000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.58660409556314,
|
| 163 |
+
"grad_norm": 0.3306196630001068,
|
| 164 |
+
"learning_rate": 4.266744880546075e-08,
|
| 165 |
+
"loss": 0.307,
|
| 166 |
+
"step": 5500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.58660409556314,
|
| 170 |
+
"eval_loss": 1.0982334613800049,
|
| 171 |
+
"eval_runtime": 62.4882,
|
| 172 |
+
"eval_samples_per_second": 15.795,
|
| 173 |
+
"eval_steps_per_second": 1.984,
|
| 174 |
+
"step": 5500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.6399317406143344,
|
| 178 |
+
"grad_norm": 149.38189697265625,
|
| 179 |
+
"learning_rate": 4.2000853242320816e-08,
|
| 180 |
+
"loss": 0.2638,
|
| 181 |
+
"step": 6000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.6399317406143344,
|
| 185 |
+
"eval_loss": 1.0779541730880737,
|
| 186 |
+
"eval_runtime": 63.4101,
|
| 187 |
+
"eval_samples_per_second": 15.565,
|
| 188 |
+
"eval_steps_per_second": 1.956,
|
| 189 |
+
"step": 6000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.693259385665529,
|
| 193 |
+
"grad_norm": 353.0036315917969,
|
| 194 |
+
"learning_rate": 4.1334257679180886e-08,
|
| 195 |
+
"loss": 0.2226,
|
| 196 |
+
"step": 6500
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.693259385665529,
|
| 200 |
+
"eval_loss": 1.0613036155700684,
|
| 201 |
+
"eval_runtime": 62.8292,
|
| 202 |
+
"eval_samples_per_second": 15.709,
|
| 203 |
+
"eval_steps_per_second": 1.974,
|
| 204 |
+
"step": 6500
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.7465870307167235,
|
| 208 |
+
"grad_norm": 0.13188917934894562,
|
| 209 |
+
"learning_rate": 4.0667662116040955e-08,
|
| 210 |
+
"loss": 0.2453,
|
| 211 |
+
"step": 7000
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.7465870307167235,
|
| 215 |
+
"eval_loss": 1.0443507432937622,
|
| 216 |
+
"eval_runtime": 62.7457,
|
| 217 |
+
"eval_samples_per_second": 15.73,
|
| 218 |
+
"eval_steps_per_second": 1.976,
|
| 219 |
+
"step": 7000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.7999146757679181,
|
| 223 |
+
"grad_norm": 59.391639709472656,
|
| 224 |
+
"learning_rate": 4.000106655290102e-08,
|
| 225 |
+
"loss": 0.272,
|
| 226 |
+
"step": 7500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.7999146757679181,
|
| 230 |
+
"eval_loss": 1.0300519466400146,
|
| 231 |
+
"eval_runtime": 62.3719,
|
| 232 |
+
"eval_samples_per_second": 15.824,
|
| 233 |
+
"eval_steps_per_second": 1.988,
|
| 234 |
+
"step": 7500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.8532423208191127,
|
| 238 |
+
"grad_norm": 495.4797668457031,
|
| 239 |
+
"learning_rate": 3.9334470989761093e-08,
|
| 240 |
+
"loss": 0.283,
|
| 241 |
+
"step": 8000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.8532423208191127,
|
| 245 |
+
"eval_loss": 1.0166898965835571,
|
| 246 |
+
"eval_runtime": 62.0338,
|
| 247 |
+
"eval_samples_per_second": 15.911,
|
| 248 |
+
"eval_steps_per_second": 1.999,
|
| 249 |
+
"step": 8000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.9065699658703071,
|
| 253 |
+
"grad_norm": 159.0310821533203,
|
| 254 |
+
"learning_rate": 3.866787542662116e-08,
|
| 255 |
+
"loss": 0.2331,
|
| 256 |
+
"step": 8500
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.9065699658703071,
|
| 260 |
+
"eval_loss": 1.003544807434082,
|
| 261 |
+
"eval_runtime": 62.9704,
|
| 262 |
+
"eval_samples_per_second": 15.674,
|
| 263 |
+
"eval_steps_per_second": 1.969,
|
| 264 |
+
"step": 8500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.9598976109215017,
|
| 268 |
+
"grad_norm": 45.40327835083008,
|
| 269 |
+
"learning_rate": 3.8001279863481225e-08,
|
| 270 |
+
"loss": 0.2362,
|
| 271 |
+
"step": 9000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.9598976109215017,
|
| 275 |
+
"eval_loss": 0.992453396320343,
|
| 276 |
+
"eval_runtime": 63.1405,
|
| 277 |
+
"eval_samples_per_second": 15.632,
|
| 278 |
+
"eval_steps_per_second": 1.964,
|
| 279 |
+
"step": 9000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.0132252559726962,
|
| 283 |
+
"grad_norm": 621.8538818359375,
|
| 284 |
+
"learning_rate": 3.7334684300341295e-08,
|
| 285 |
+
"loss": 0.2396,
|
| 286 |
+
"step": 9500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 1.0132252559726962,
|
| 290 |
+
"eval_loss": 0.9829725027084351,
|
| 291 |
+
"eval_runtime": 62.3244,
|
| 292 |
+
"eval_samples_per_second": 15.836,
|
| 293 |
+
"eval_steps_per_second": 1.99,
|
| 294 |
+
"step": 9500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.0665529010238908,
|
| 298 |
+
"grad_norm": 461.60736083984375,
|
| 299 |
+
"learning_rate": 3.6668088737201364e-08,
|
| 300 |
+
"loss": 0.2013,
|
| 301 |
+
"step": 10000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 1.0665529010238908,
|
| 305 |
+
"eval_loss": 0.9735796451568604,
|
| 306 |
+
"eval_runtime": 63.0763,
|
| 307 |
+
"eval_samples_per_second": 15.648,
|
| 308 |
+
"eval_steps_per_second": 1.966,
|
| 309 |
+
"step": 10000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.1198805460750854,
|
| 313 |
+
"grad_norm": 7.0309906005859375,
|
| 314 |
+
"learning_rate": 3.6001493174061433e-08,
|
| 315 |
+
"loss": 0.2082,
|
| 316 |
+
"step": 10500
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.1198805460750854,
|
| 320 |
+
"eval_loss": 0.9639086127281189,
|
| 321 |
+
"eval_runtime": 63.0238,
|
| 322 |
+
"eval_samples_per_second": 15.661,
|
| 323 |
+
"eval_steps_per_second": 1.968,
|
| 324 |
+
"step": 10500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.17320819112628,
|
| 328 |
+
"grad_norm": 304.01995849609375,
|
| 329 |
+
"learning_rate": 3.5334897610921496e-08,
|
| 330 |
+
"loss": 0.2023,
|
| 331 |
+
"step": 11000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.17320819112628,
|
| 335 |
+
"eval_loss": 0.9558469653129578,
|
| 336 |
+
"eval_runtime": 62.1866,
|
| 337 |
+
"eval_samples_per_second": 15.872,
|
| 338 |
+
"eval_steps_per_second": 1.994,
|
| 339 |
+
"step": 11000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.2265358361774745,
|
| 343 |
+
"grad_norm": 0.0017981340643018484,
|
| 344 |
+
"learning_rate": 3.4668302047781565e-08,
|
| 345 |
+
"loss": 0.2331,
|
| 346 |
+
"step": 11500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.2265358361774745,
|
| 350 |
+
"eval_loss": 0.9465070962905884,
|
| 351 |
+
"eval_runtime": 63.1547,
|
| 352 |
+
"eval_samples_per_second": 15.628,
|
| 353 |
+
"eval_steps_per_second": 1.963,
|
| 354 |
+
"step": 11500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.2798634812286689,
|
| 358 |
+
"grad_norm": 0.8147886395454407,
|
| 359 |
+
"learning_rate": 3.4001706484641635e-08,
|
| 360 |
+
"loss": 0.1784,
|
| 361 |
+
"step": 12000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 1.2798634812286689,
|
| 365 |
+
"eval_loss": 0.9392343163490295,
|
| 366 |
+
"eval_runtime": 62.9221,
|
| 367 |
+
"eval_samples_per_second": 15.686,
|
| 368 |
+
"eval_steps_per_second": 1.971,
|
| 369 |
+
"step": 12000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 1.3331911262798635,
|
| 373 |
+
"grad_norm": 0.01276802271604538,
|
| 374 |
+
"learning_rate": 3.3335110921501704e-08,
|
| 375 |
+
"loss": 0.1953,
|
| 376 |
+
"step": 12500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 1.3331911262798635,
|
| 380 |
+
"eval_loss": 0.9316345453262329,
|
| 381 |
+
"eval_runtime": 62.2816,
|
| 382 |
+
"eval_samples_per_second": 15.847,
|
| 383 |
+
"eval_steps_per_second": 1.991,
|
| 384 |
+
"step": 12500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.386518771331058,
|
| 388 |
+
"grad_norm": 0.07341516762971878,
|
| 389 |
+
"learning_rate": 3.2668515358361773e-08,
|
| 390 |
+
"loss": 0.1867,
|
| 391 |
+
"step": 13000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 1.386518771331058,
|
| 395 |
+
"eval_loss": 0.9270105957984924,
|
| 396 |
+
"eval_runtime": 63.2619,
|
| 397 |
+
"eval_samples_per_second": 15.602,
|
| 398 |
+
"eval_steps_per_second": 1.96,
|
| 399 |
+
"step": 13000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 1.4398464163822526,
|
| 403 |
+
"grad_norm": 17.55596923828125,
|
| 404 |
+
"learning_rate": 3.200191979522184e-08,
|
| 405 |
+
"loss": 0.22,
|
| 406 |
+
"step": 13500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 1.4398464163822526,
|
| 410 |
+
"eval_loss": 0.919725775718689,
|
| 411 |
+
"eval_runtime": 63.3459,
|
| 412 |
+
"eval_samples_per_second": 15.581,
|
| 413 |
+
"eval_steps_per_second": 1.958,
|
| 414 |
+
"step": 13500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 1.493174061433447,
|
| 418 |
+
"grad_norm": 3.4780771732330322,
|
| 419 |
+
"learning_rate": 3.133532423208191e-08,
|
| 420 |
+
"loss": 0.1656,
|
| 421 |
+
"step": 14000
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 1.493174061433447,
|
| 425 |
+
"eval_loss": 0.9148014783859253,
|
| 426 |
+
"eval_runtime": 62.257,
|
| 427 |
+
"eval_samples_per_second": 15.854,
|
| 428 |
+
"eval_steps_per_second": 1.992,
|
| 429 |
+
"step": 14000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.5465017064846416,
|
| 433 |
+
"grad_norm": 158.22390747070312,
|
| 434 |
+
"learning_rate": 3.066872866894198e-08,
|
| 435 |
+
"loss": 0.1968,
|
| 436 |
+
"step": 14500
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.5465017064846416,
|
| 440 |
+
"eval_loss": 0.9096349477767944,
|
| 441 |
+
"eval_runtime": 63.3671,
|
| 442 |
+
"eval_samples_per_second": 15.576,
|
| 443 |
+
"eval_steps_per_second": 1.957,
|
| 444 |
+
"step": 14500
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.5998293515358362,
|
| 448 |
+
"grad_norm": 621.3922729492188,
|
| 449 |
+
"learning_rate": 3.0002133105802044e-08,
|
| 450 |
+
"loss": 0.1676,
|
| 451 |
+
"step": 15000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.5998293515358362,
|
| 455 |
+
"eval_loss": 0.9056846499443054,
|
| 456 |
+
"eval_runtime": 63.2369,
|
| 457 |
+
"eval_samples_per_second": 15.608,
|
| 458 |
+
"eval_steps_per_second": 1.961,
|
| 459 |
+
"step": 15000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 1.6531569965870307,
|
| 463 |
+
"grad_norm": 26.23398208618164,
|
| 464 |
+
"learning_rate": 2.9335537542662113e-08,
|
| 465 |
+
"loss": 0.2074,
|
| 466 |
+
"step": 15500
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 1.6531569965870307,
|
| 470 |
+
"eval_loss": 0.8993550539016724,
|
| 471 |
+
"eval_runtime": 62.3767,
|
| 472 |
+
"eval_samples_per_second": 15.823,
|
| 473 |
+
"eval_steps_per_second": 1.988,
|
| 474 |
+
"step": 15500
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 1.7064846416382253,
|
| 478 |
+
"grad_norm": 283.3314514160156,
|
| 479 |
+
"learning_rate": 2.8668941979522183e-08,
|
| 480 |
+
"loss": 0.1847,
|
| 481 |
+
"step": 16000
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 1.7064846416382253,
|
| 485 |
+
"eval_loss": 0.8954482078552246,
|
| 486 |
+
"eval_runtime": 63.2168,
|
| 487 |
+
"eval_samples_per_second": 15.613,
|
| 488 |
+
"eval_steps_per_second": 1.962,
|
| 489 |
+
"step": 16000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 1.75981228668942,
|
| 493 |
+
"grad_norm": 12.151030540466309,
|
| 494 |
+
"learning_rate": 2.800234641638225e-08,
|
| 495 |
+
"loss": 0.1845,
|
| 496 |
+
"step": 16500
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 1.75981228668942,
|
| 500 |
+
"eval_loss": 0.8900338411331177,
|
| 501 |
+
"eval_runtime": 63.5581,
|
| 502 |
+
"eval_samples_per_second": 15.529,
|
| 503 |
+
"eval_steps_per_second": 1.951,
|
| 504 |
+
"step": 16500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.8131399317406145,
|
| 508 |
+
"grad_norm": 0.7503092288970947,
|
| 509 |
+
"learning_rate": 2.7335750853242318e-08,
|
| 510 |
+
"loss": 0.1721,
|
| 511 |
+
"step": 17000
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 1.8131399317406145,
|
| 515 |
+
"eval_loss": 0.8872589468955994,
|
| 516 |
+
"eval_runtime": 62.0282,
|
| 517 |
+
"eval_samples_per_second": 15.912,
|
| 518 |
+
"eval_steps_per_second": 1.999,
|
| 519 |
+
"step": 17000
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 1.8664675767918089,
|
| 523 |
+
"grad_norm": 180.9355926513672,
|
| 524 |
+
"learning_rate": 2.6669155290102384e-08,
|
| 525 |
+
"loss": 0.2627,
|
| 526 |
+
"step": 17500
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 1.8664675767918089,
|
| 530 |
+
"eval_loss": 0.8810189962387085,
|
| 531 |
+
"eval_runtime": 62.7212,
|
| 532 |
+
"eval_samples_per_second": 15.736,
|
| 533 |
+
"eval_steps_per_second": 1.977,
|
| 534 |
+
"step": 17500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.9197952218430034,
|
| 538 |
+
"grad_norm": 0.10503419488668442,
|
| 539 |
+
"learning_rate": 2.600255972696246e-08,
|
| 540 |
+
"loss": 0.1623,
|
| 541 |
+
"step": 18000
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.9197952218430034,
|
| 545 |
+
"eval_loss": 0.8773773908615112,
|
| 546 |
+
"eval_runtime": 62.9181,
|
| 547 |
+
"eval_samples_per_second": 15.687,
|
| 548 |
+
"eval_steps_per_second": 1.971,
|
| 549 |
+
"step": 18000
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.9731228668941978,
|
| 553 |
+
"grad_norm": 477.76715087890625,
|
| 554 |
+
"learning_rate": 2.5335964163822526e-08,
|
| 555 |
+
"loss": 0.2162,
|
| 556 |
+
"step": 18500
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 1.9731228668941978,
|
| 560 |
+
"eval_loss": 0.8713328242301941,
|
| 561 |
+
"eval_runtime": 62.3477,
|
| 562 |
+
"eval_samples_per_second": 15.831,
|
| 563 |
+
"eval_steps_per_second": 1.989,
|
| 564 |
+
"step": 18500
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.0264505119453924,
|
| 568 |
+
"grad_norm": 222.06802368164062,
|
| 569 |
+
"learning_rate": 2.4669368600682595e-08,
|
| 570 |
+
"loss": 0.1802,
|
| 571 |
+
"step": 19000
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.0264505119453924,
|
| 575 |
+
"eval_loss": 0.8679063320159912,
|
| 576 |
+
"eval_runtime": 63.0768,
|
| 577 |
+
"eval_samples_per_second": 15.648,
|
| 578 |
+
"eval_steps_per_second": 1.966,
|
| 579 |
+
"step": 19000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 2.079778156996587,
|
| 583 |
+
"grad_norm": 48.63727569580078,
|
| 584 |
+
"learning_rate": 2.400277303754266e-08,
|
| 585 |
+
"loss": 0.179,
|
| 586 |
+
"step": 19500
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 2.079778156996587,
|
| 590 |
+
"eval_loss": 0.86326664686203,
|
| 591 |
+
"eval_runtime": 63.3068,
|
| 592 |
+
"eval_samples_per_second": 15.591,
|
| 593 |
+
"eval_steps_per_second": 1.959,
|
| 594 |
+
"step": 19500
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 2.1331058020477816,
|
| 598 |
+
"grad_norm": 0.5827552080154419,
|
| 599 |
+
"learning_rate": 2.333617747440273e-08,
|
| 600 |
+
"loss": 0.1549,
|
| 601 |
+
"step": 20000
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 2.1331058020477816,
|
| 605 |
+
"eval_loss": 0.8606277108192444,
|
| 606 |
+
"eval_runtime": 62.3116,
|
| 607 |
+
"eval_samples_per_second": 15.84,
|
| 608 |
+
"eval_steps_per_second": 1.99,
|
| 609 |
+
"step": 20000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 2.186433447098976,
|
| 613 |
+
"grad_norm": 0.060927845537662506,
|
| 614 |
+
"learning_rate": 2.2669581911262796e-08,
|
| 615 |
+
"loss": 0.1742,
|
| 616 |
+
"step": 20500
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 2.186433447098976,
|
| 620 |
+
"eval_loss": 0.8584741353988647,
|
| 621 |
+
"eval_runtime": 63.1659,
|
| 622 |
+
"eval_samples_per_second": 15.626,
|
| 623 |
+
"eval_steps_per_second": 1.963,
|
| 624 |
+
"step": 20500
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 2.2397610921501707,
|
| 628 |
+
"grad_norm": 16.044307708740234,
|
| 629 |
+
"learning_rate": 2.2002986348122866e-08,
|
| 630 |
+
"loss": 0.1448,
|
| 631 |
+
"step": 21000
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 2.2397610921501707,
|
| 635 |
+
"eval_loss": 0.8546220064163208,
|
| 636 |
+
"eval_runtime": 63.0133,
|
| 637 |
+
"eval_samples_per_second": 15.663,
|
| 638 |
+
"eval_steps_per_second": 1.968,
|
| 639 |
+
"step": 21000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 2.2930887372013653,
|
| 643 |
+
"grad_norm": 0.07948703318834305,
|
| 644 |
+
"learning_rate": 2.1336390784982935e-08,
|
| 645 |
+
"loss": 0.2066,
|
| 646 |
+
"step": 21500
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 2.2930887372013653,
|
| 650 |
+
"eval_loss": 0.8512936234474182,
|
| 651 |
+
"eval_runtime": 62.3887,
|
| 652 |
+
"eval_samples_per_second": 15.82,
|
| 653 |
+
"eval_steps_per_second": 1.988,
|
| 654 |
+
"step": 21500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 2.34641638225256,
|
| 658 |
+
"grad_norm": 0.0013235179940238595,
|
| 659 |
+
"learning_rate": 2.0669795221843004e-08,
|
| 660 |
+
"loss": 0.1483,
|
| 661 |
+
"step": 22000
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 2.34641638225256,
|
| 665 |
+
"eval_loss": 0.8480631709098816,
|
| 666 |
+
"eval_runtime": 63.1614,
|
| 667 |
+
"eval_samples_per_second": 15.627,
|
| 668 |
+
"eval_steps_per_second": 1.963,
|
| 669 |
+
"step": 22000
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 2.399744027303754,
|
| 673 |
+
"grad_norm": 0.001996780978515744,
|
| 674 |
+
"learning_rate": 2.000319965870307e-08,
|
| 675 |
+
"loss": 0.1813,
|
| 676 |
+
"step": 22500
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 2.399744027303754,
|
| 680 |
+
"eval_loss": 0.84471195936203,
|
| 681 |
+
"eval_runtime": 63.0032,
|
| 682 |
+
"eval_samples_per_second": 15.666,
|
| 683 |
+
"eval_steps_per_second": 1.968,
|
| 684 |
+
"step": 22500
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 2.453071672354949,
|
| 688 |
+
"grad_norm": 0.24369017779827118,
|
| 689 |
+
"learning_rate": 1.933660409556314e-08,
|
| 690 |
+
"loss": 0.1617,
|
| 691 |
+
"step": 23000
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 2.453071672354949,
|
| 695 |
+
"eval_loss": 0.8411402702331543,
|
| 696 |
+
"eval_runtime": 62.707,
|
| 697 |
+
"eval_samples_per_second": 15.74,
|
| 698 |
+
"eval_steps_per_second": 1.977,
|
| 699 |
+
"step": 23000
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 2.506399317406143,
|
| 703 |
+
"grad_norm": 9.377041816711426,
|
| 704 |
+
"learning_rate": 1.8670008532423206e-08,
|
| 705 |
+
"loss": 0.1664,
|
| 706 |
+
"step": 23500
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 2.506399317406143,
|
| 710 |
+
"eval_loss": 0.8393682837486267,
|
| 711 |
+
"eval_runtime": 63.143,
|
| 712 |
+
"eval_samples_per_second": 15.631,
|
| 713 |
+
"eval_steps_per_second": 1.964,
|
| 714 |
+
"step": 23500
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 2.5597269624573378,
|
| 718 |
+
"grad_norm": 118.065673828125,
|
| 719 |
+
"learning_rate": 1.800341296928328e-08,
|
| 720 |
+
"loss": 0.1786,
|
| 721 |
+
"step": 24000
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 2.5597269624573378,
|
| 725 |
+
"eval_loss": 0.8357825875282288,
|
| 726 |
+
"eval_runtime": 63.3183,
|
| 727 |
+
"eval_samples_per_second": 15.588,
|
| 728 |
+
"eval_steps_per_second": 1.958,
|
| 729 |
+
"step": 24000
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 2.6130546075085324,
|
| 733 |
+
"grad_norm": 3.0652036666870117,
|
| 734 |
+
"learning_rate": 1.7336817406143344e-08,
|
| 735 |
+
"loss": 0.1465,
|
| 736 |
+
"step": 24500
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 2.6130546075085324,
|
| 740 |
+
"eval_loss": 0.8330481648445129,
|
| 741 |
+
"eval_runtime": 63.0565,
|
| 742 |
+
"eval_samples_per_second": 15.653,
|
| 743 |
+
"eval_steps_per_second": 1.966,
|
| 744 |
+
"step": 24500
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 2.666382252559727,
|
| 748 |
+
"grad_norm": 68.4100570678711,
|
| 749 |
+
"learning_rate": 1.6670221843003414e-08,
|
| 750 |
+
"loss": 0.1289,
|
| 751 |
+
"step": 25000
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 2.666382252559727,
|
| 755 |
+
"eval_loss": 0.831382691860199,
|
| 756 |
+
"eval_runtime": 62.8466,
|
| 757 |
+
"eval_samples_per_second": 15.705,
|
| 758 |
+
"eval_steps_per_second": 1.973,
|
| 759 |
+
"step": 25000
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 2.7197098976109215,
|
| 763 |
+
"grad_norm": 0.03573409095406532,
|
| 764 |
+
"learning_rate": 1.600362627986348e-08,
|
| 765 |
+
"loss": 0.1662,
|
| 766 |
+
"step": 25500
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 2.7197098976109215,
|
| 770 |
+
"eval_loss": 0.829609751701355,
|
| 771 |
+
"eval_runtime": 63.536,
|
| 772 |
+
"eval_samples_per_second": 15.535,
|
| 773 |
+
"eval_steps_per_second": 1.952,
|
| 774 |
+
"step": 25500
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 2.773037542662116,
|
| 778 |
+
"grad_norm": 85.1174087524414,
|
| 779 |
+
"learning_rate": 1.533703071672355e-08,
|
| 780 |
+
"loss": 0.1463,
|
| 781 |
+
"step": 26000
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 2.773037542662116,
|
| 785 |
+
"eval_loss": 0.8262338638305664,
|
| 786 |
+
"eval_runtime": 63.1492,
|
| 787 |
+
"eval_samples_per_second": 15.63,
|
| 788 |
+
"eval_steps_per_second": 1.964,
|
| 789 |
+
"step": 26000
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 2.8263651877133107,
|
| 793 |
+
"grad_norm": 1.988323450088501,
|
| 794 |
+
"learning_rate": 1.4670435153583615e-08,
|
| 795 |
+
"loss": 0.1471,
|
| 796 |
+
"step": 26500
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 2.8263651877133107,
|
| 800 |
+
"eval_loss": 0.8249030709266663,
|
| 801 |
+
"eval_runtime": 63.1063,
|
| 802 |
+
"eval_samples_per_second": 15.64,
|
| 803 |
+
"eval_steps_per_second": 1.965,
|
| 804 |
+
"step": 26500
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 2.8796928327645053,
|
| 808 |
+
"grad_norm": 14.553828239440918,
|
| 809 |
+
"learning_rate": 1.4003839590443686e-08,
|
| 810 |
+
"loss": 0.167,
|
| 811 |
+
"step": 27000
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 2.8796928327645053,
|
| 815 |
+
"eval_loss": 0.821887731552124,
|
| 816 |
+
"eval_runtime": 63.5261,
|
| 817 |
+
"eval_samples_per_second": 15.537,
|
| 818 |
+
"eval_steps_per_second": 1.952,
|
| 819 |
+
"step": 27000
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 2.9330204778157,
|
| 823 |
+
"grad_norm": 0.4691038131713867,
|
| 824 |
+
"learning_rate": 1.3337244027303754e-08,
|
| 825 |
+
"loss": 0.1268,
|
| 826 |
+
"step": 27500
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 2.9330204778157,
|
| 830 |
+
"eval_loss": 0.8203587532043457,
|
| 831 |
+
"eval_runtime": 63.3176,
|
| 832 |
+
"eval_samples_per_second": 15.588,
|
| 833 |
+
"eval_steps_per_second": 1.958,
|
| 834 |
+
"step": 27500
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 2.986348122866894,
|
| 838 |
+
"grad_norm": 327.5917663574219,
|
| 839 |
+
"learning_rate": 1.2670648464163821e-08,
|
| 840 |
+
"loss": 0.177,
|
| 841 |
+
"step": 28000
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 2.986348122866894,
|
| 845 |
+
"eval_loss": 0.8176783919334412,
|
| 846 |
+
"eval_runtime": 63.0051,
|
| 847 |
+
"eval_samples_per_second": 15.665,
|
| 848 |
+
"eval_steps_per_second": 1.968,
|
| 849 |
+
"step": 28000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 3.0396757679180886,
|
| 853 |
+
"grad_norm": 1.4327431472338503e-06,
|
| 854 |
+
"learning_rate": 1.200405290102389e-08,
|
| 855 |
+
"loss": 0.1206,
|
| 856 |
+
"step": 28500
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 3.0396757679180886,
|
| 860 |
+
"eval_loss": 0.8165684938430786,
|
| 861 |
+
"eval_runtime": 63.4667,
|
| 862 |
+
"eval_samples_per_second": 15.551,
|
| 863 |
+
"eval_steps_per_second": 1.954,
|
| 864 |
+
"step": 28500
|
| 865 |
+
}
|
| 866 |
+
],
|
| 867 |
+
"logging_steps": 500,
|
| 868 |
+
"max_steps": 37504,
|
| 869 |
+
"num_input_tokens_seen": 0,
|
| 870 |
+
"num_train_epochs": 4,
|
| 871 |
+
"save_steps": 500,
|
| 872 |
+
"stateful_callbacks": {
|
| 873 |
+
"TrainerControl": {
|
| 874 |
+
"args": {
|
| 875 |
+
"should_epoch_stop": false,
|
| 876 |
+
"should_evaluate": false,
|
| 877 |
+
"should_log": false,
|
| 878 |
+
"should_save": true,
|
| 879 |
+
"should_training_stop": false
|
| 880 |
+
},
|
| 881 |
+
"attributes": {}
|
| 882 |
+
}
|
| 883 |
+
},
|
| 884 |
+
"total_flos": 1.025383252761252e+16,
|
| 885 |
+
"train_batch_size": 2,
|
| 886 |
+
"trial_name": null,
|
| 887 |
+
"trial_params": null
|
| 888 |
+
}
|
checkpoint-28500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:87e35dca972e13467c750106c19301b098fbd2918fda1087487e227bac3aca37
|
| 3 |
+
size 4847
|