🍻 cheers
Browse files- README.md +4 -3
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Apr20_09-53-02_49f104673c10/events.out.tfevents.1713668078.49f104673c10.914.4 +3 -0
- train_results.json +8 -0
- trainer_state.json +778 -0
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: facebook/convnext-base-224-22k
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
metrics:
|
| 7 |
- accuracy
|
|
@@ -15,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# idbwbase
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [facebook/convnext-base-224-22k](https://huggingface.co/facebook/convnext-base-224-22k) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
-
- Accuracy: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: facebook/convnext-base-224-22k
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
|
|
|
| 16 |
|
| 17 |
# idbwbase
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [facebook/convnext-base-224-22k](https://huggingface.co/facebook/convnext-base-224-22k) on the indian_food_images dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.0978
|
| 22 |
+
- Accuracy: 0.9741
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9741072939583686,
|
| 4 |
+
"eval_loss": 0.097813181579113,
|
| 5 |
+
"eval_runtime": 360.845,
|
| 6 |
+
"eval_samples_per_second": 147.379,
|
| 7 |
+
"eval_steps_per_second": 18.423,
|
| 8 |
+
"total_flos": 2.383420595495315e+20,
|
| 9 |
+
"train_loss": 0.22288281881624433,
|
| 10 |
+
"train_runtime": 60913.5069,
|
| 11 |
+
"train_samples_per_second": 49.473,
|
| 12 |
+
"train_steps_per_second": 0.773
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9741072939583686,
|
| 4 |
+
"eval_loss": 0.097813181579113,
|
| 5 |
+
"eval_runtime": 360.845,
|
| 6 |
+
"eval_samples_per_second": 147.379,
|
| 7 |
+
"eval_steps_per_second": 18.423
|
| 8 |
+
}
|
runs/Apr20_09-53-02_49f104673c10/events.out.tfevents.1713668078.49f104673c10.914.4
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5da082219aaf1eac918c92e9380a9ca6677a8c5aeb48d7756bb6046399ce4f2a
|
| 3 |
+
size 417
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 2.383420595495315e+20,
|
| 4 |
+
"train_loss": 0.22288281881624433,
|
| 5 |
+
"train_runtime": 60913.5069,
|
| 6 |
+
"train_samples_per_second": 49.473,
|
| 7 |
+
"train_steps_per_second": 0.773
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,778 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.097813181579113,
|
| 3 |
+
"best_model_checkpoint": "idbwbase/checkpoint-42381",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 47090,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.11,
|
| 13 |
+
"grad_norm": 3.36433744430542,
|
| 14 |
+
"learning_rate": 0.0001978764068804417,
|
| 15 |
+
"loss": 0.4791,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.21,
|
| 20 |
+
"grad_norm": 3.1407039165496826,
|
| 21 |
+
"learning_rate": 0.0001957528137608834,
|
| 22 |
+
"loss": 0.454,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.32,
|
| 27 |
+
"grad_norm": 2.483966112136841,
|
| 28 |
+
"learning_rate": 0.00019362922064132513,
|
| 29 |
+
"loss": 0.4364,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.42,
|
| 34 |
+
"grad_norm": 2.059077501296997,
|
| 35 |
+
"learning_rate": 0.00019150987470800598,
|
| 36 |
+
"loss": 0.4149,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.53,
|
| 41 |
+
"grad_norm": 2.1167092323303223,
|
| 42 |
+
"learning_rate": 0.00018938628158844767,
|
| 43 |
+
"loss": 0.41,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.64,
|
| 48 |
+
"grad_norm": 1.9980089664459229,
|
| 49 |
+
"learning_rate": 0.00018726268846888937,
|
| 50 |
+
"loss": 0.3953,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.74,
|
| 55 |
+
"grad_norm": 1.5476139783859253,
|
| 56 |
+
"learning_rate": 0.00018513909534933107,
|
| 57 |
+
"loss": 0.3892,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.85,
|
| 62 |
+
"grad_norm": 1.8769443035125732,
|
| 63 |
+
"learning_rate": 0.0001830197494160119,
|
| 64 |
+
"loss": 0.3826,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.96,
|
| 69 |
+
"grad_norm": 2.0475404262542725,
|
| 70 |
+
"learning_rate": 0.0001808961562964536,
|
| 71 |
+
"loss": 0.373,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.0,
|
| 76 |
+
"eval_accuracy": 0.870536469791843,
|
| 77 |
+
"eval_loss": 0.30019834637641907,
|
| 78 |
+
"eval_runtime": 363.8544,
|
| 79 |
+
"eval_samples_per_second": 146.16,
|
| 80 |
+
"eval_steps_per_second": 18.271,
|
| 81 |
+
"step": 4709
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.06,
|
| 85 |
+
"grad_norm": 3.492917776107788,
|
| 86 |
+
"learning_rate": 0.0001787725631768953,
|
| 87 |
+
"loss": 0.3639,
|
| 88 |
+
"step": 5000
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.17,
|
| 92 |
+
"grad_norm": 3.2921090126037598,
|
| 93 |
+
"learning_rate": 0.00017664897005733704,
|
| 94 |
+
"loss": 0.3553,
|
| 95 |
+
"step": 5500
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.27,
|
| 99 |
+
"grad_norm": 2.0384316444396973,
|
| 100 |
+
"learning_rate": 0.00017452537693777873,
|
| 101 |
+
"loss": 0.3499,
|
| 102 |
+
"step": 6000
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.38,
|
| 106 |
+
"grad_norm": 2.055048942565918,
|
| 107 |
+
"learning_rate": 0.00017240178381822043,
|
| 108 |
+
"loss": 0.3506,
|
| 109 |
+
"step": 6500
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.49,
|
| 113 |
+
"grad_norm": 1.965047001838684,
|
| 114 |
+
"learning_rate": 0.00017027819069866215,
|
| 115 |
+
"loss": 0.3423,
|
| 116 |
+
"step": 7000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.59,
|
| 120 |
+
"grad_norm": 2.3037126064300537,
|
| 121 |
+
"learning_rate": 0.00016815459757910385,
|
| 122 |
+
"loss": 0.3402,
|
| 123 |
+
"step": 7500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.7,
|
| 127 |
+
"grad_norm": 2.647536516189575,
|
| 128 |
+
"learning_rate": 0.00016603100445954555,
|
| 129 |
+
"loss": 0.3323,
|
| 130 |
+
"step": 8000
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.81,
|
| 134 |
+
"grad_norm": 2.767207384109497,
|
| 135 |
+
"learning_rate": 0.0001639116585262264,
|
| 136 |
+
"loss": 0.3294,
|
| 137 |
+
"step": 8500
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.91,
|
| 141 |
+
"grad_norm": 2.31170654296875,
|
| 142 |
+
"learning_rate": 0.00016179231259290722,
|
| 143 |
+
"loss": 0.3244,
|
| 144 |
+
"step": 9000
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 2.0,
|
| 148 |
+
"eval_accuracy": 0.9044395554803407,
|
| 149 |
+
"eval_loss": 0.22620604932308197,
|
| 150 |
+
"eval_runtime": 365.6085,
|
| 151 |
+
"eval_samples_per_second": 145.459,
|
| 152 |
+
"eval_steps_per_second": 18.183,
|
| 153 |
+
"step": 9418
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.02,
|
| 157 |
+
"grad_norm": 1.986091136932373,
|
| 158 |
+
"learning_rate": 0.00015966871947334892,
|
| 159 |
+
"loss": 0.3175,
|
| 160 |
+
"step": 9500
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 2.12,
|
| 164 |
+
"grad_norm": 2.7250773906707764,
|
| 165 |
+
"learning_rate": 0.0001575451263537906,
|
| 166 |
+
"loss": 0.3038,
|
| 167 |
+
"step": 10000
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 2.23,
|
| 171 |
+
"grad_norm": 2.0195140838623047,
|
| 172 |
+
"learning_rate": 0.0001554215332342323,
|
| 173 |
+
"loss": 0.3037,
|
| 174 |
+
"step": 10500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 2.34,
|
| 178 |
+
"grad_norm": 2.9471521377563477,
|
| 179 |
+
"learning_rate": 0.00015329794011467404,
|
| 180 |
+
"loss": 0.2911,
|
| 181 |
+
"step": 11000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 2.44,
|
| 185 |
+
"grad_norm": 1.8268228769302368,
|
| 186 |
+
"learning_rate": 0.00015117434699511573,
|
| 187 |
+
"loss": 0.2931,
|
| 188 |
+
"step": 11500
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.55,
|
| 192 |
+
"grad_norm": 2.0443902015686035,
|
| 193 |
+
"learning_rate": 0.00014905075387555746,
|
| 194 |
+
"loss": 0.2903,
|
| 195 |
+
"step": 12000
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.65,
|
| 199 |
+
"grad_norm": 2.8994696140289307,
|
| 200 |
+
"learning_rate": 0.00014692716075599915,
|
| 201 |
+
"loss": 0.2864,
|
| 202 |
+
"step": 12500
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 2.76,
|
| 206 |
+
"grad_norm": 2.1983399391174316,
|
| 207 |
+
"learning_rate": 0.00014480781482267998,
|
| 208 |
+
"loss": 0.2833,
|
| 209 |
+
"step": 13000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 2.87,
|
| 213 |
+
"grad_norm": 2.4135706424713135,
|
| 214 |
+
"learning_rate": 0.0001426884688893608,
|
| 215 |
+
"loss": 0.2788,
|
| 216 |
+
"step": 13500
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 2.97,
|
| 220 |
+
"grad_norm": 1.9159047603607178,
|
| 221 |
+
"learning_rate": 0.0001405648757698025,
|
| 222 |
+
"loss": 0.2801,
|
| 223 |
+
"step": 14000
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 3.0,
|
| 227 |
+
"eval_accuracy": 0.9195953442018766,
|
| 228 |
+
"eval_loss": 0.1986866444349289,
|
| 229 |
+
"eval_runtime": 365.1106,
|
| 230 |
+
"eval_samples_per_second": 145.657,
|
| 231 |
+
"eval_steps_per_second": 18.208,
|
| 232 |
+
"step": 14127
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 3.08,
|
| 236 |
+
"grad_norm": 2.0817108154296875,
|
| 237 |
+
"learning_rate": 0.00013844128265024422,
|
| 238 |
+
"loss": 0.2584,
|
| 239 |
+
"step": 14500
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 3.19,
|
| 243 |
+
"grad_norm": 2.8534085750579834,
|
| 244 |
+
"learning_rate": 0.00013631768953068594,
|
| 245 |
+
"loss": 0.2559,
|
| 246 |
+
"step": 15000
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 3.29,
|
| 250 |
+
"grad_norm": 3.98779559135437,
|
| 251 |
+
"learning_rate": 0.00013419409641112764,
|
| 252 |
+
"loss": 0.2511,
|
| 253 |
+
"step": 15500
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 3.4,
|
| 257 |
+
"grad_norm": 3.104475975036621,
|
| 258 |
+
"learning_rate": 0.00013207475047780846,
|
| 259 |
+
"loss": 0.2517,
|
| 260 |
+
"step": 16000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 3.5,
|
| 264 |
+
"grad_norm": 2.7032172679901123,
|
| 265 |
+
"learning_rate": 0.00012995115735825016,
|
| 266 |
+
"loss": 0.2504,
|
| 267 |
+
"step": 16500
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 3.61,
|
| 271 |
+
"grad_norm": 3.3657124042510986,
|
| 272 |
+
"learning_rate": 0.00012782756423869186,
|
| 273 |
+
"loss": 0.2461,
|
| 274 |
+
"step": 17000
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 3.72,
|
| 278 |
+
"grad_norm": 2.80549693107605,
|
| 279 |
+
"learning_rate": 0.00012570397111913358,
|
| 280 |
+
"loss": 0.2451,
|
| 281 |
+
"step": 17500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 3.82,
|
| 285 |
+
"grad_norm": 1.6164408922195435,
|
| 286 |
+
"learning_rate": 0.0001235803779995753,
|
| 287 |
+
"loss": 0.2421,
|
| 288 |
+
"step": 18000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 3.93,
|
| 292 |
+
"grad_norm": 2.589157819747925,
|
| 293 |
+
"learning_rate": 0.00012145678488001699,
|
| 294 |
+
"loss": 0.2366,
|
| 295 |
+
"step": 18500
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 4.0,
|
| 299 |
+
"eval_accuracy": 0.9344502735939527,
|
| 300 |
+
"eval_loss": 0.17884854972362518,
|
| 301 |
+
"eval_runtime": 360.8393,
|
| 302 |
+
"eval_samples_per_second": 147.381,
|
| 303 |
+
"eval_steps_per_second": 18.424,
|
| 304 |
+
"step": 18836
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 4.03,
|
| 308 |
+
"grad_norm": 2.659468412399292,
|
| 309 |
+
"learning_rate": 0.0001193331917604587,
|
| 310 |
+
"loss": 0.2287,
|
| 311 |
+
"step": 19000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 4.14,
|
| 315 |
+
"grad_norm": 2.452038526535034,
|
| 316 |
+
"learning_rate": 0.00011720959864090042,
|
| 317 |
+
"loss": 0.2205,
|
| 318 |
+
"step": 19500
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 4.25,
|
| 322 |
+
"grad_norm": 3.7109389305114746,
|
| 323 |
+
"learning_rate": 0.00011509025270758123,
|
| 324 |
+
"loss": 0.2076,
|
| 325 |
+
"step": 20000
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 4.35,
|
| 329 |
+
"grad_norm": 2.5269851684570312,
|
| 330 |
+
"learning_rate": 0.00011296665958802296,
|
| 331 |
+
"loss": 0.2134,
|
| 332 |
+
"step": 20500
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 4.46,
|
| 336 |
+
"grad_norm": 2.7143073081970215,
|
| 337 |
+
"learning_rate": 0.00011084306646846465,
|
| 338 |
+
"loss": 0.2144,
|
| 339 |
+
"step": 21000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 4.57,
|
| 343 |
+
"grad_norm": 3.1681575775146484,
|
| 344 |
+
"learning_rate": 0.00010871947334890635,
|
| 345 |
+
"loss": 0.2091,
|
| 346 |
+
"step": 21500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 4.67,
|
| 350 |
+
"grad_norm": 1.2098946571350098,
|
| 351 |
+
"learning_rate": 0.00010659588022934806,
|
| 352 |
+
"loss": 0.2065,
|
| 353 |
+
"step": 22000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 4.78,
|
| 357 |
+
"grad_norm": 1.6173245906829834,
|
| 358 |
+
"learning_rate": 0.00010447228710978979,
|
| 359 |
+
"loss": 0.2107,
|
| 360 |
+
"step": 22500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 4.88,
|
| 364 |
+
"grad_norm": 4.180131435394287,
|
| 365 |
+
"learning_rate": 0.00010235294117647058,
|
| 366 |
+
"loss": 0.2053,
|
| 367 |
+
"step": 23000
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 4.99,
|
| 371 |
+
"grad_norm": 1.9634019136428833,
|
| 372 |
+
"learning_rate": 0.0001002293480569123,
|
| 373 |
+
"loss": 0.2051,
|
| 374 |
+
"step": 23500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 5.0,
|
| 378 |
+
"eval_accuracy": 0.9483650175814671,
|
| 379 |
+
"eval_loss": 0.1462855190038681,
|
| 380 |
+
"eval_runtime": 362.8319,
|
| 381 |
+
"eval_samples_per_second": 146.572,
|
| 382 |
+
"eval_steps_per_second": 18.323,
|
| 383 |
+
"step": 23545
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 5.1,
|
| 387 |
+
"grad_norm": 2.8337090015411377,
|
| 388 |
+
"learning_rate": 9.810575493735401e-05,
|
| 389 |
+
"loss": 0.1896,
|
| 390 |
+
"step": 24000
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 5.2,
|
| 394 |
+
"grad_norm": 1.7976356744766235,
|
| 395 |
+
"learning_rate": 9.598216181779571e-05,
|
| 396 |
+
"loss": 0.1872,
|
| 397 |
+
"step": 24500
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 5.31,
|
| 401 |
+
"grad_norm": 2.647800922393799,
|
| 402 |
+
"learning_rate": 9.385856869823742e-05,
|
| 403 |
+
"loss": 0.1819,
|
| 404 |
+
"step": 25000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 5.42,
|
| 408 |
+
"grad_norm": 2.1115944385528564,
|
| 409 |
+
"learning_rate": 9.173922276491824e-05,
|
| 410 |
+
"loss": 0.1837,
|
| 411 |
+
"step": 25500
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 5.52,
|
| 415 |
+
"grad_norm": 3.922607183456421,
|
| 416 |
+
"learning_rate": 8.961562964535995e-05,
|
| 417 |
+
"loss": 0.1868,
|
| 418 |
+
"step": 26000
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 5.63,
|
| 422 |
+
"grad_norm": 1.7281802892684937,
|
| 423 |
+
"learning_rate": 8.749203652580165e-05,
|
| 424 |
+
"loss": 0.1832,
|
| 425 |
+
"step": 26500
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 5.73,
|
| 429 |
+
"grad_norm": 2.4881973266601562,
|
| 430 |
+
"learning_rate": 8.536844340624338e-05,
|
| 431 |
+
"loss": 0.1791,
|
| 432 |
+
"step": 27000
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 5.84,
|
| 436 |
+
"grad_norm": 1.2277300357818604,
|
| 437 |
+
"learning_rate": 8.324909747292418e-05,
|
| 438 |
+
"loss": 0.1805,
|
| 439 |
+
"step": 27500
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 5.95,
|
| 443 |
+
"grad_norm": 1.151065468788147,
|
| 444 |
+
"learning_rate": 8.112975153960502e-05,
|
| 445 |
+
"loss": 0.1764,
|
| 446 |
+
"step": 28000
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 6.0,
|
| 450 |
+
"eval_accuracy": 0.9593087756905662,
|
| 451 |
+
"eval_loss": 0.1202462688088417,
|
| 452 |
+
"eval_runtime": 359.7308,
|
| 453 |
+
"eval_samples_per_second": 147.836,
|
| 454 |
+
"eval_steps_per_second": 18.48,
|
| 455 |
+
"step": 28254
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 6.05,
|
| 459 |
+
"grad_norm": 2.2972559928894043,
|
| 460 |
+
"learning_rate": 7.900615842004673e-05,
|
| 461 |
+
"loss": 0.1694,
|
| 462 |
+
"step": 28500
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 6.16,
|
| 466 |
+
"grad_norm": 1.6081655025482178,
|
| 467 |
+
"learning_rate": 7.688256530048843e-05,
|
| 468 |
+
"loss": 0.1656,
|
| 469 |
+
"step": 29000
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 6.26,
|
| 473 |
+
"grad_norm": 4.132796764373779,
|
| 474 |
+
"learning_rate": 7.475897218093014e-05,
|
| 475 |
+
"loss": 0.1589,
|
| 476 |
+
"step": 29500
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 6.37,
|
| 480 |
+
"grad_norm": 2.355618476867676,
|
| 481 |
+
"learning_rate": 7.263537906137185e-05,
|
| 482 |
+
"loss": 0.1618,
|
| 483 |
+
"step": 30000
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 6.48,
|
| 487 |
+
"grad_norm": 2.159538745880127,
|
| 488 |
+
"learning_rate": 7.051178594181356e-05,
|
| 489 |
+
"loss": 0.1572,
|
| 490 |
+
"step": 30500
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 6.58,
|
| 494 |
+
"grad_norm": 3.115480661392212,
|
| 495 |
+
"learning_rate": 6.838819282225526e-05,
|
| 496 |
+
"loss": 0.163,
|
| 497 |
+
"step": 31000
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 6.69,
|
| 501 |
+
"grad_norm": 2.2360424995422363,
|
| 502 |
+
"learning_rate": 6.626459970269697e-05,
|
| 503 |
+
"loss": 0.1607,
|
| 504 |
+
"step": 31500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 6.8,
|
| 508 |
+
"grad_norm": 2.970757246017456,
|
| 509 |
+
"learning_rate": 6.414525376937779e-05,
|
| 510 |
+
"loss": 0.1599,
|
| 511 |
+
"step": 32000
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 6.9,
|
| 515 |
+
"grad_norm": 6.401217460632324,
|
| 516 |
+
"learning_rate": 6.20216606498195e-05,
|
| 517 |
+
"loss": 0.1595,
|
| 518 |
+
"step": 32500
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 7.0,
|
| 522 |
+
"eval_accuracy": 0.9654763919444914,
|
| 523 |
+
"eval_loss": 0.12429220974445343,
|
| 524 |
+
"eval_runtime": 360.8363,
|
| 525 |
+
"eval_samples_per_second": 147.383,
|
| 526 |
+
"eval_steps_per_second": 18.424,
|
| 527 |
+
"step": 32963
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 7.01,
|
| 531 |
+
"grad_norm": 3.2262797355651855,
|
| 532 |
+
"learning_rate": 5.9898067530261204e-05,
|
| 533 |
+
"loss": 0.1528,
|
| 534 |
+
"step": 33000
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 7.11,
|
| 538 |
+
"grad_norm": 3.578483819961548,
|
| 539 |
+
"learning_rate": 5.7774474410702914e-05,
|
| 540 |
+
"loss": 0.1473,
|
| 541 |
+
"step": 33500
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 7.22,
|
| 545 |
+
"grad_norm": 3.9829201698303223,
|
| 546 |
+
"learning_rate": 5.565088129114462e-05,
|
| 547 |
+
"loss": 0.1441,
|
| 548 |
+
"step": 34000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 7.33,
|
| 552 |
+
"grad_norm": 3.7269842624664307,
|
| 553 |
+
"learning_rate": 5.3531535357825446e-05,
|
| 554 |
+
"loss": 0.1385,
|
| 555 |
+
"step": 34500
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 7.43,
|
| 559 |
+
"grad_norm": 2.6239781379699707,
|
| 560 |
+
"learning_rate": 5.140794223826715e-05,
|
| 561 |
+
"loss": 0.1406,
|
| 562 |
+
"step": 35000
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 7.54,
|
| 566 |
+
"grad_norm": 1.6224650144577026,
|
| 567 |
+
"learning_rate": 4.9284349118708855e-05,
|
| 568 |
+
"loss": 0.1429,
|
| 569 |
+
"step": 35500
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 7.64,
|
| 573 |
+
"grad_norm": 5.391174793243408,
|
| 574 |
+
"learning_rate": 4.7160755999150565e-05,
|
| 575 |
+
"loss": 0.1457,
|
| 576 |
+
"step": 36000
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 7.75,
|
| 580 |
+
"grad_norm": 5.336463928222656,
|
| 581 |
+
"learning_rate": 4.504141006583139e-05,
|
| 582 |
+
"loss": 0.1359,
|
| 583 |
+
"step": 36500
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 7.86,
|
| 587 |
+
"grad_norm": 3.005934238433838,
|
| 588 |
+
"learning_rate": 4.29178169462731e-05,
|
| 589 |
+
"loss": 0.1417,
|
| 590 |
+
"step": 37000
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 7.96,
|
| 594 |
+
"grad_norm": 3.7323451042175293,
|
| 595 |
+
"learning_rate": 4.079422382671481e-05,
|
| 596 |
+
"loss": 0.1359,
|
| 597 |
+
"step": 37500
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 8.0,
|
| 601 |
+
"eval_accuracy": 0.9658712698144074,
|
| 602 |
+
"eval_loss": 0.11875798553228378,
|
| 603 |
+
"eval_runtime": 368.3038,
|
| 604 |
+
"eval_samples_per_second": 144.394,
|
| 605 |
+
"eval_steps_per_second": 18.05,
|
| 606 |
+
"step": 37672
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 8.07,
|
| 610 |
+
"grad_norm": 2.7839584350585938,
|
| 611 |
+
"learning_rate": 3.867063070715651e-05,
|
| 612 |
+
"loss": 0.1335,
|
| 613 |
+
"step": 38000
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 8.18,
|
| 617 |
+
"grad_norm": 4.733059406280518,
|
| 618 |
+
"learning_rate": 3.6551284773837333e-05,
|
| 619 |
+
"loss": 0.1313,
|
| 620 |
+
"step": 38500
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 8.28,
|
| 624 |
+
"grad_norm": 2.712228536605835,
|
| 625 |
+
"learning_rate": 3.4427691654279044e-05,
|
| 626 |
+
"loss": 0.1325,
|
| 627 |
+
"step": 39000
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 8.39,
|
| 631 |
+
"grad_norm": 4.026779651641846,
|
| 632 |
+
"learning_rate": 3.230409853472075e-05,
|
| 633 |
+
"loss": 0.1272,
|
| 634 |
+
"step": 39500
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 8.49,
|
| 638 |
+
"grad_norm": 3.8425519466400146,
|
| 639 |
+
"learning_rate": 3.018050541516246e-05,
|
| 640 |
+
"loss": 0.1307,
|
| 641 |
+
"step": 40000
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 8.6,
|
| 645 |
+
"grad_norm": 3.5189483165740967,
|
| 646 |
+
"learning_rate": 2.806115948184328e-05,
|
| 647 |
+
"loss": 0.129,
|
| 648 |
+
"step": 40500
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 8.71,
|
| 652 |
+
"grad_norm": 4.5053486824035645,
|
| 653 |
+
"learning_rate": 2.5937566362284988e-05,
|
| 654 |
+
"loss": 0.1257,
|
| 655 |
+
"step": 41000
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 8.81,
|
| 659 |
+
"grad_norm": 4.372600078582764,
|
| 660 |
+
"learning_rate": 2.3818220428965812e-05,
|
| 661 |
+
"loss": 0.1238,
|
| 662 |
+
"step": 41500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 8.92,
|
| 666 |
+
"grad_norm": 1.8681597709655762,
|
| 667 |
+
"learning_rate": 2.169462730940752e-05,
|
| 668 |
+
"loss": 0.1231,
|
| 669 |
+
"step": 42000
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 9.0,
|
| 673 |
+
"eval_accuracy": 0.9741072939583686,
|
| 674 |
+
"eval_loss": 0.097813181579113,
|
| 675 |
+
"eval_runtime": 359.2297,
|
| 676 |
+
"eval_samples_per_second": 148.042,
|
| 677 |
+
"eval_steps_per_second": 18.506,
|
| 678 |
+
"step": 42381
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 9.03,
|
| 682 |
+
"grad_norm": 3.127023220062256,
|
| 683 |
+
"learning_rate": 1.9571034189849227e-05,
|
| 684 |
+
"loss": 0.1249,
|
| 685 |
+
"step": 42500
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 9.13,
|
| 689 |
+
"grad_norm": 1.650771141052246,
|
| 690 |
+
"learning_rate": 1.744744107029093e-05,
|
| 691 |
+
"loss": 0.118,
|
| 692 |
+
"step": 43000
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 9.24,
|
| 696 |
+
"grad_norm": 3.3975789546966553,
|
| 697 |
+
"learning_rate": 1.532384795073264e-05,
|
| 698 |
+
"loss": 0.1198,
|
| 699 |
+
"step": 43500
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 9.34,
|
| 703 |
+
"grad_norm": 1.8186984062194824,
|
| 704 |
+
"learning_rate": 1.3200254831174346e-05,
|
| 705 |
+
"loss": 0.1167,
|
| 706 |
+
"step": 44000
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 9.45,
|
| 710 |
+
"grad_norm": 7.110922336578369,
|
| 711 |
+
"learning_rate": 1.1076661711616054e-05,
|
| 712 |
+
"loss": 0.1198,
|
| 713 |
+
"step": 44500
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 9.56,
|
| 717 |
+
"grad_norm": 6.389682292938232,
|
| 718 |
+
"learning_rate": 8.953068592057761e-06,
|
| 719 |
+
"loss": 0.1187,
|
| 720 |
+
"step": 45000
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 9.66,
|
| 724 |
+
"grad_norm": 2.857072353363037,
|
| 725 |
+
"learning_rate": 6.833722658738587e-06,
|
| 726 |
+
"loss": 0.1129,
|
| 727 |
+
"step": 45500
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 9.77,
|
| 731 |
+
"grad_norm": 0.7192772626876831,
|
| 732 |
+
"learning_rate": 4.710129539180293e-06,
|
| 733 |
+
"loss": 0.1133,
|
| 734 |
+
"step": 46000
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 9.87,
|
| 738 |
+
"grad_norm": 3.4914393424987793,
|
| 739 |
+
"learning_rate": 2.5865364196220007e-06,
|
| 740 |
+
"loss": 0.1151,
|
| 741 |
+
"step": 46500
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 9.98,
|
| 745 |
+
"grad_norm": 1.8187050819396973,
|
| 746 |
+
"learning_rate": 4.671904863028244e-07,
|
| 747 |
+
"loss": 0.1162,
|
| 748 |
+
"step": 47000
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 10.0,
|
| 752 |
+
"eval_accuracy": 0.9761004870160396,
|
| 753 |
+
"eval_loss": 0.10011033713817596,
|
| 754 |
+
"eval_runtime": 359.0659,
|
| 755 |
+
"eval_samples_per_second": 148.109,
|
| 756 |
+
"eval_steps_per_second": 18.515,
|
| 757 |
+
"step": 47090
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 10.0,
|
| 761 |
+
"step": 47090,
|
| 762 |
+
"total_flos": 2.383420595495315e+20,
|
| 763 |
+
"train_loss": 0.22288281881624433,
|
| 764 |
+
"train_runtime": 60913.5069,
|
| 765 |
+
"train_samples_per_second": 49.473,
|
| 766 |
+
"train_steps_per_second": 0.773
|
| 767 |
+
}
|
| 768 |
+
],
|
| 769 |
+
"logging_steps": 500,
|
| 770 |
+
"max_steps": 47090,
|
| 771 |
+
"num_input_tokens_seen": 0,
|
| 772 |
+
"num_train_epochs": 10,
|
| 773 |
+
"save_steps": 500,
|
| 774 |
+
"total_flos": 2.383420595495315e+20,
|
| 775 |
+
"train_batch_size": 64,
|
| 776 |
+
"trial_name": null,
|
| 777 |
+
"trial_params": null
|
| 778 |
+
}
|