End of training
Browse files- README.md +3 -3
- all_results.json +16 -0
- eval_results.json +10 -0
- train_results.json +9 -0
- trainer_state.json +733 -0
README.md
CHANGED
|
@@ -15,10 +15,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# tinyllama-1.1b-3T-bees-internal-2048-vN
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [pszemraj/tinyllama-1.1b-3T](https://huggingface.co/pszemraj/tinyllama-1.1b-3T) on the
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 2.
|
| 21 |
-
- Accuracy: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 15 |
|
| 16 |
# tinyllama-1.1b-3T-bees-internal-2048-vN
|
| 17 |
|
| 18 |
+
This model is a fine-tuned version of [pszemraj/tinyllama-1.1b-3T](https://huggingface.co/pszemraj/tinyllama-1.1b-3T) on the BEE-spoke-data/bees-internal dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 2.1640
|
| 21 |
+
- Accuracy: 0.5406
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"eval_accuracy": 0.5406350757205667,
|
| 4 |
+
"eval_loss": 2.1639890670776367,
|
| 5 |
+
"eval_runtime": 14.336,
|
| 6 |
+
"eval_samples": 300,
|
| 7 |
+
"eval_samples_per_second": 20.926,
|
| 8 |
+
"eval_steps_per_second": 10.463,
|
| 9 |
+
"perplexity": 8.705796490465838,
|
| 10 |
+
"train_loss": 2.223597173579037,
|
| 11 |
+
"train_runtime": 8040.0291,
|
| 12 |
+
"train_samples": 16416,
|
| 13 |
+
"train_samples_per_second": 4.084,
|
| 14 |
+
"train_steps_per_second": 0.064,
|
| 15 |
+
"train_tokens_per_second": 8363.146
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"eval_accuracy": 0.5406350757205667,
|
| 4 |
+
"eval_loss": 2.1639890670776367,
|
| 5 |
+
"eval_runtime": 14.336,
|
| 6 |
+
"eval_samples": 300,
|
| 7 |
+
"eval_samples_per_second": 20.926,
|
| 8 |
+
"eval_steps_per_second": 10.463,
|
| 9 |
+
"perplexity": 8.705796490465838
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"train_loss": 2.223597173579037,
|
| 4 |
+
"train_runtime": 8040.0291,
|
| 5 |
+
"train_samples": 16416,
|
| 6 |
+
"train_samples_per_second": 4.084,
|
| 7 |
+
"train_steps_per_second": 0.064,
|
| 8 |
+
"train_tokens_per_second": 8363.146
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,733 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.996101364522417,
|
| 5 |
+
"eval_steps": 50,
|
| 6 |
+
"global_step": 512,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"learning_rate": 1.923076923076923e-05,
|
| 14 |
+
"loss": 2.7497,
|
| 15 |
+
"step": 5
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.04,
|
| 19 |
+
"learning_rate": 3.846153846153846e-05,
|
| 20 |
+
"loss": 2.7052,
|
| 21 |
+
"step": 10
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.06,
|
| 25 |
+
"learning_rate": 5.769230769230769e-05,
|
| 26 |
+
"loss": 2.6213,
|
| 27 |
+
"step": 15
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.08,
|
| 31 |
+
"learning_rate": 7.692307692307693e-05,
|
| 32 |
+
"loss": 2.6018,
|
| 33 |
+
"step": 20
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.1,
|
| 37 |
+
"learning_rate": 9.615384615384617e-05,
|
| 38 |
+
"loss": 2.5872,
|
| 39 |
+
"step": 25
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.12,
|
| 43 |
+
"learning_rate": 9.998328666948438e-05,
|
| 44 |
+
"loss": 2.5665,
|
| 45 |
+
"step": 30
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.14,
|
| 49 |
+
"learning_rate": 9.991540791356342e-05,
|
| 50 |
+
"loss": 2.5494,
|
| 51 |
+
"step": 35
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.16,
|
| 55 |
+
"learning_rate": 9.979538999730047e-05,
|
| 56 |
+
"loss": 2.5143,
|
| 57 |
+
"step": 40
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.18,
|
| 61 |
+
"learning_rate": 9.962335828546048e-05,
|
| 62 |
+
"loss": 2.441,
|
| 63 |
+
"step": 45
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.19,
|
| 67 |
+
"learning_rate": 9.939949247384046e-05,
|
| 68 |
+
"loss": 2.4432,
|
| 69 |
+
"step": 50
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.19,
|
| 73 |
+
"eval_accuracy": 0.5032714541605602,
|
| 74 |
+
"eval_loss": 2.3850491046905518,
|
| 75 |
+
"eval_runtime": 14.1442,
|
| 76 |
+
"eval_samples_per_second": 21.21,
|
| 77 |
+
"eval_steps_per_second": 10.605,
|
| 78 |
+
"step": 50
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.21,
|
| 82 |
+
"learning_rate": 9.912402640156811e-05,
|
| 83 |
+
"loss": 2.4841,
|
| 84 |
+
"step": 55
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.23,
|
| 88 |
+
"learning_rate": 9.879724780684519e-05,
|
| 89 |
+
"loss": 2.4121,
|
| 90 |
+
"step": 60
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.25,
|
| 94 |
+
"learning_rate": 9.84194980263903e-05,
|
| 95 |
+
"loss": 2.4448,
|
| 96 |
+
"step": 65
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.27,
|
| 100 |
+
"learning_rate": 9.799117163889559e-05,
|
| 101 |
+
"loss": 2.4297,
|
| 102 |
+
"step": 70
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.29,
|
| 106 |
+
"learning_rate": 9.751271605286941e-05,
|
| 107 |
+
"loss": 2.4581,
|
| 108 |
+
"step": 75
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.31,
|
| 112 |
+
"learning_rate": 9.698463103929542e-05,
|
| 113 |
+
"loss": 2.435,
|
| 114 |
+
"step": 80
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.33,
|
| 118 |
+
"learning_rate": 9.640746820959684e-05,
|
| 119 |
+
"loss": 2.3842,
|
| 120 |
+
"step": 85
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.35,
|
| 124 |
+
"learning_rate": 9.57818304394503e-05,
|
| 125 |
+
"loss": 2.3872,
|
| 126 |
+
"step": 90
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.37,
|
| 130 |
+
"learning_rate": 9.51083712390519e-05,
|
| 131 |
+
"loss": 2.426,
|
| 132 |
+
"step": 95
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.39,
|
| 136 |
+
"learning_rate": 9.438779407049281e-05,
|
| 137 |
+
"loss": 2.3655,
|
| 138 |
+
"step": 100
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.39,
|
| 142 |
+
"eval_accuracy": 0.5129278619117408,
|
| 143 |
+
"eval_loss": 2.3123531341552734,
|
| 144 |
+
"eval_runtime": 14.1415,
|
| 145 |
+
"eval_samples_per_second": 21.214,
|
| 146 |
+
"eval_steps_per_second": 10.607,
|
| 147 |
+
"step": 100
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.41,
|
| 151 |
+
"learning_rate": 9.362085161295769e-05,
|
| 152 |
+
"loss": 2.402,
|
| 153 |
+
"step": 105
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.43,
|
| 157 |
+
"learning_rate": 9.280834497651334e-05,
|
| 158 |
+
"loss": 2.3523,
|
| 159 |
+
"step": 110
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.45,
|
| 163 |
+
"learning_rate": 9.195112286530873e-05,
|
| 164 |
+
"loss": 2.3354,
|
| 165 |
+
"step": 115
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.47,
|
| 169 |
+
"learning_rate": 9.105008069106093e-05,
|
| 170 |
+
"loss": 2.37,
|
| 171 |
+
"step": 120
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.49,
|
| 175 |
+
"learning_rate": 9.01061596377522e-05,
|
| 176 |
+
"loss": 2.3597,
|
| 177 |
+
"step": 125
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.51,
|
| 181 |
+
"learning_rate": 8.912034567851599e-05,
|
| 182 |
+
"loss": 2.3646,
|
| 183 |
+
"step": 130
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 0.53,
|
| 187 |
+
"learning_rate": 8.809366854573831e-05,
|
| 188 |
+
"loss": 2.3601,
|
| 189 |
+
"step": 135
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.55,
|
| 193 |
+
"learning_rate": 8.702720065545024e-05,
|
| 194 |
+
"loss": 2.3424,
|
| 195 |
+
"step": 140
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.57,
|
| 199 |
+
"learning_rate": 8.592205598713539e-05,
|
| 200 |
+
"loss": 2.3362,
|
| 201 |
+
"step": 145
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.58,
|
| 205 |
+
"learning_rate": 8.47793889201221e-05,
|
| 206 |
+
"loss": 2.374,
|
| 207 |
+
"step": 150
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.58,
|
| 211 |
+
"eval_accuracy": 0.5214769581501384,
|
| 212 |
+
"eval_loss": 2.258837938308716,
|
| 213 |
+
"eval_runtime": 14.3574,
|
| 214 |
+
"eval_samples_per_second": 20.895,
|
| 215 |
+
"eval_steps_per_second": 10.448,
|
| 216 |
+
"step": 150
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.6,
|
| 220 |
+
"learning_rate": 8.360039302777612e-05,
|
| 221 |
+
"loss": 2.3383,
|
| 222 |
+
"step": 155
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 0.62,
|
| 226 |
+
"learning_rate": 8.238629983075294e-05,
|
| 227 |
+
"loss": 2.319,
|
| 228 |
+
"step": 160
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 0.64,
|
| 232 |
+
"learning_rate": 8.113837751061246e-05,
|
| 233 |
+
"loss": 2.3517,
|
| 234 |
+
"step": 165
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.66,
|
| 238 |
+
"learning_rate": 7.985792958513931e-05,
|
| 239 |
+
"loss": 2.307,
|
| 240 |
+
"step": 170
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.68,
|
| 244 |
+
"learning_rate": 7.854629354675291e-05,
|
| 245 |
+
"loss": 2.2915,
|
| 246 |
+
"step": 175
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 0.7,
|
| 250 |
+
"learning_rate": 7.720483946542914e-05,
|
| 251 |
+
"loss": 2.3198,
|
| 252 |
+
"step": 180
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 0.72,
|
| 256 |
+
"learning_rate": 7.583496855759316e-05,
|
| 257 |
+
"loss": 2.3304,
|
| 258 |
+
"step": 185
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 0.74,
|
| 262 |
+
"learning_rate": 7.443811172247821e-05,
|
| 263 |
+
"loss": 2.2926,
|
| 264 |
+
"step": 190
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.76,
|
| 268 |
+
"learning_rate": 7.30157280474793e-05,
|
| 269 |
+
"loss": 2.2753,
|
| 270 |
+
"step": 195
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 0.78,
|
| 274 |
+
"learning_rate": 7.156930328406268e-05,
|
| 275 |
+
"loss": 2.3558,
|
| 276 |
+
"step": 200
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.78,
|
| 280 |
+
"eval_accuracy": 0.5291418335775933,
|
| 281 |
+
"eval_loss": 2.2132091522216797,
|
| 282 |
+
"eval_runtime": 14.0805,
|
| 283 |
+
"eval_samples_per_second": 21.306,
|
| 284 |
+
"eval_steps_per_second": 10.653,
|
| 285 |
+
"step": 200
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.8,
|
| 289 |
+
"learning_rate": 7.01003482958237e-05,
|
| 290 |
+
"loss": 2.3026,
|
| 291 |
+
"step": 205
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 0.82,
|
| 295 |
+
"learning_rate": 6.861039748031351e-05,
|
| 296 |
+
"loss": 2.3327,
|
| 297 |
+
"step": 210
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.84,
|
| 301 |
+
"learning_rate": 6.710100716628344e-05,
|
| 302 |
+
"loss": 2.3047,
|
| 303 |
+
"step": 215
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.86,
|
| 307 |
+
"learning_rate": 6.557375398802123e-05,
|
| 308 |
+
"loss": 2.2849,
|
| 309 |
+
"step": 220
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.88,
|
| 313 |
+
"learning_rate": 6.403023323847695e-05,
|
| 314 |
+
"loss": 2.2972,
|
| 315 |
+
"step": 225
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 0.9,
|
| 319 |
+
"learning_rate": 6.247205720289907e-05,
|
| 320 |
+
"loss": 2.3069,
|
| 321 |
+
"step": 230
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.92,
|
| 325 |
+
"learning_rate": 6.09008534747213e-05,
|
| 326 |
+
"loss": 2.3305,
|
| 327 |
+
"step": 235
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 0.94,
|
| 331 |
+
"learning_rate": 5.9318263255459116e-05,
|
| 332 |
+
"loss": 2.2845,
|
| 333 |
+
"step": 240
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 0.96,
|
| 337 |
+
"learning_rate": 5.772593964039203e-05,
|
| 338 |
+
"loss": 2.2698,
|
| 339 |
+
"step": 245
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.97,
|
| 343 |
+
"learning_rate": 5.6125545891822274e-05,
|
| 344 |
+
"loss": 2.2677,
|
| 345 |
+
"step": 250
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.97,
|
| 349 |
+
"eval_accuracy": 0.5348330890734408,
|
| 350 |
+
"eval_loss": 2.182823896408081,
|
| 351 |
+
"eval_runtime": 14.0998,
|
| 352 |
+
"eval_samples_per_second": 21.277,
|
| 353 |
+
"eval_steps_per_second": 10.638,
|
| 354 |
+
"step": 250
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.99,
|
| 358 |
+
"learning_rate": 5.451875370171341e-05,
|
| 359 |
+
"loss": 2.2776,
|
| 360 |
+
"step": 255
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.01,
|
| 364 |
+
"learning_rate": 5.290724144552379e-05,
|
| 365 |
+
"loss": 2.1158,
|
| 366 |
+
"step": 260
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.03,
|
| 370 |
+
"learning_rate": 5.129269242905882e-05,
|
| 371 |
+
"loss": 2.0758,
|
| 372 |
+
"step": 265
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 1.05,
|
| 376 |
+
"learning_rate": 4.967679313017303e-05,
|
| 377 |
+
"loss": 2.053,
|
| 378 |
+
"step": 270
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 1.07,
|
| 382 |
+
"learning_rate": 4.806123143715916e-05,
|
| 383 |
+
"loss": 2.0824,
|
| 384 |
+
"step": 275
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.09,
|
| 388 |
+
"learning_rate": 4.6447694885663514e-05,
|
| 389 |
+
"loss": 2.062,
|
| 390 |
+
"step": 280
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 1.11,
|
| 394 |
+
"learning_rate": 4.4837868895969936e-05,
|
| 395 |
+
"loss": 2.0986,
|
| 396 |
+
"step": 285
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 1.13,
|
| 400 |
+
"learning_rate": 4.323343501249346e-05,
|
| 401 |
+
"loss": 2.0749,
|
| 402 |
+
"step": 290
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.15,
|
| 406 |
+
"learning_rate": 4.1636069147322246e-05,
|
| 407 |
+
"loss": 2.0365,
|
| 408 |
+
"step": 295
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.17,
|
| 412 |
+
"learning_rate": 4.004743982964298e-05,
|
| 413 |
+
"loss": 2.0701,
|
| 414 |
+
"step": 300
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 1.17,
|
| 418 |
+
"eval_accuracy": 0.5372854583943983,
|
| 419 |
+
"eval_loss": 2.178772211074829,
|
| 420 |
+
"eval_runtime": 14.087,
|
| 421 |
+
"eval_samples_per_second": 21.296,
|
| 422 |
+
"eval_steps_per_second": 10.648,
|
| 423 |
+
"step": 300
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.19,
|
| 427 |
+
"learning_rate": 3.846920646287799e-05,
|
| 428 |
+
"loss": 2.0554,
|
| 429 |
+
"step": 305
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.21,
|
| 433 |
+
"learning_rate": 3.6903017591354706e-05,
|
| 434 |
+
"loss": 2.0432,
|
| 435 |
+
"step": 310
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 1.23,
|
| 439 |
+
"learning_rate": 3.535050917831797e-05,
|
| 440 |
+
"loss": 2.0663,
|
| 441 |
+
"step": 315
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 1.25,
|
| 445 |
+
"learning_rate": 3.381330289708396e-05,
|
| 446 |
+
"loss": 2.0673,
|
| 447 |
+
"step": 320
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 1.27,
|
| 451 |
+
"learning_rate": 3.2293004437120624e-05,
|
| 452 |
+
"loss": 2.0106,
|
| 453 |
+
"step": 325
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 1.29,
|
| 457 |
+
"learning_rate": 3.079120182682412e-05,
|
| 458 |
+
"loss": 2.0177,
|
| 459 |
+
"step": 330
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 1.31,
|
| 463 |
+
"learning_rate": 2.9309463774743046e-05,
|
| 464 |
+
"loss": 2.0314,
|
| 465 |
+
"step": 335
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.33,
|
| 469 |
+
"learning_rate": 2.7849338030983257e-05,
|
| 470 |
+
"loss": 2.06,
|
| 471 |
+
"step": 340
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.35,
|
| 475 |
+
"learning_rate": 2.641234977050484e-05,
|
| 476 |
+
"loss": 2.0765,
|
| 477 |
+
"step": 345
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 1.36,
|
| 481 |
+
"learning_rate": 2.500000000000001e-05,
|
| 482 |
+
"loss": 2.0766,
|
| 483 |
+
"step": 350
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 1.36,
|
| 487 |
+
"eval_accuracy": 0.5397508549096238,
|
| 488 |
+
"eval_loss": 2.1673190593719482,
|
| 489 |
+
"eval_runtime": 14.2515,
|
| 490 |
+
"eval_samples_per_second": 21.05,
|
| 491 |
+
"eval_steps_per_second": 10.525,
|
| 492 |
+
"step": 350
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.38,
|
| 496 |
+
"learning_rate": 2.361376399001592e-05,
|
| 497 |
+
"loss": 2.0481,
|
| 498 |
+
"step": 355
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 1.4,
|
| 502 |
+
"learning_rate": 2.225508973396016e-05,
|
| 503 |
+
"loss": 2.0246,
|
| 504 |
+
"step": 360
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.42,
|
| 508 |
+
"learning_rate": 2.0925396435598664e-05,
|
| 509 |
+
"loss": 2.0563,
|
| 510 |
+
"step": 365
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 1.44,
|
| 514 |
+
"learning_rate": 1.9626073026625818e-05,
|
| 515 |
+
"loss": 2.0373,
|
| 516 |
+
"step": 370
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 1.46,
|
| 520 |
+
"learning_rate": 1.835847671585526e-05,
|
| 521 |
+
"loss": 2.0495,
|
| 522 |
+
"step": 375
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 1.48,
|
| 526 |
+
"learning_rate": 1.7123931571546827e-05,
|
| 527 |
+
"loss": 2.0612,
|
| 528 |
+
"step": 380
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 1.5,
|
| 532 |
+
"learning_rate": 1.592372713835055e-05,
|
| 533 |
+
"loss": 2.0543,
|
| 534 |
+
"step": 385
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.52,
|
| 538 |
+
"learning_rate": 1.4759117090312197e-05,
|
| 539 |
+
"loss": 2.0719,
|
| 540 |
+
"step": 390
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 1.54,
|
| 544 |
+
"learning_rate": 1.3631317921347563e-05,
|
| 545 |
+
"loss": 2.0468,
|
| 546 |
+
"step": 395
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 1.56,
|
| 550 |
+
"learning_rate": 1.25415076745532e-05,
|
| 551 |
+
"loss": 2.0669,
|
| 552 |
+
"step": 400
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 1.56,
|
| 556 |
+
"eval_accuracy": 0.5402182055039896,
|
| 557 |
+
"eval_loss": 2.165104866027832,
|
| 558 |
+
"eval_runtime": 14.1201,
|
| 559 |
+
"eval_samples_per_second": 21.246,
|
| 560 |
+
"eval_steps_per_second": 10.623,
|
| 561 |
+
"step": 400
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 1.58,
|
| 565 |
+
"learning_rate": 1.1490824711681025e-05,
|
| 566 |
+
"loss": 2.0491,
|
| 567 |
+
"step": 405
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 1.6,
|
| 571 |
+
"learning_rate": 1.0480366524062042e-05,
|
| 572 |
+
"loss": 2.0686,
|
| 573 |
+
"step": 410
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 1.62,
|
| 577 |
+
"learning_rate": 9.511188586221376e-06,
|
| 578 |
+
"loss": 2.0816,
|
| 579 |
+
"step": 415
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 1.64,
|
| 583 |
+
"learning_rate": 8.584303253381847e-06,
|
| 584 |
+
"loss": 2.0808,
|
| 585 |
+
"step": 420
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 1.66,
|
| 589 |
+
"learning_rate": 7.700678704007947e-06,
|
| 590 |
+
"loss": 2.0244,
|
| 591 |
+
"step": 425
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 1.68,
|
| 595 |
+
"learning_rate": 6.861237928494579e-06,
|
| 596 |
+
"loss": 2.0715,
|
| 597 |
+
"step": 430
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.7,
|
| 601 |
+
"learning_rate": 6.066857765057055e-06,
|
| 602 |
+
"loss": 2.0629,
|
| 603 |
+
"step": 435
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 1.72,
|
| 607 |
+
"learning_rate": 5.318367983829392e-06,
|
| 608 |
+
"loss": 2.0738,
|
| 609 |
+
"step": 440
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 1.73,
|
| 613 |
+
"learning_rate": 4.616550420127563e-06,
|
| 614 |
+
"loss": 2.0449,
|
| 615 |
+
"step": 445
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 1.75,
|
| 619 |
+
"learning_rate": 3.962138157783085e-06,
|
| 620 |
+
"loss": 2.0314,
|
| 621 |
+
"step": 450
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 1.75,
|
| 625 |
+
"eval_accuracy": 0.5405943657384791,
|
| 626 |
+
"eval_loss": 2.1640877723693848,
|
| 627 |
+
"eval_runtime": 14.1591,
|
| 628 |
+
"eval_samples_per_second": 21.188,
|
| 629 |
+
"eval_steps_per_second": 10.594,
|
| 630 |
+
"step": 450
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 1.77,
|
| 634 |
+
"learning_rate": 3.3558147633999728e-06,
|
| 635 |
+
"loss": 2.0818,
|
| 636 |
+
"step": 455
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 1.79,
|
| 640 |
+
"learning_rate": 2.798213572335001e-06,
|
| 641 |
+
"loss": 2.0991,
|
| 642 |
+
"step": 460
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 1.81,
|
| 646 |
+
"learning_rate": 2.2899170271469428e-06,
|
| 647 |
+
"loss": 2.0317,
|
| 648 |
+
"step": 465
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 1.83,
|
| 652 |
+
"learning_rate": 1.8314560692059835e-06,
|
| 653 |
+
"loss": 2.0518,
|
| 654 |
+
"step": 470
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.85,
|
| 658 |
+
"learning_rate": 1.4233095840986753e-06,
|
| 659 |
+
"loss": 2.013,
|
| 660 |
+
"step": 475
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.87,
|
| 664 |
+
"learning_rate": 1.0659039014077944e-06,
|
| 665 |
+
"loss": 2.0317,
|
| 666 |
+
"step": 480
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 1.89,
|
| 670 |
+
"learning_rate": 7.596123493895991e-07,
|
| 671 |
+
"loss": 2.038,
|
| 672 |
+
"step": 485
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 1.91,
|
| 676 |
+
"learning_rate": 5.047548650136513e-07,
|
| 677 |
+
"loss": 2.0414,
|
| 678 |
+
"step": 490
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 1.93,
|
| 682 |
+
"learning_rate": 3.0159765977250673e-07,
|
| 683 |
+
"loss": 2.0509,
|
| 684 |
+
"step": 495
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 1.95,
|
| 688 |
+
"learning_rate": 1.503529416103988e-07,
|
| 689 |
+
"loss": 2.0281,
|
| 690 |
+
"step": 500
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 1.95,
|
| 694 |
+
"eval_accuracy": 0.5406627585083863,
|
| 695 |
+
"eval_loss": 2.163918972015381,
|
| 696 |
+
"eval_runtime": 14.1546,
|
| 697 |
+
"eval_samples_per_second": 21.194,
|
| 698 |
+
"eval_steps_per_second": 10.597,
|
| 699 |
+
"step": 500
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 1.97,
|
| 703 |
+
"learning_rate": 5.11786932613223e-08,
|
| 704 |
+
"loss": 2.0471,
|
| 705 |
+
"step": 505
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 1.99,
|
| 709 |
+
"learning_rate": 4.178507228136397e-09,
|
| 710 |
+
"loss": 2.055,
|
| 711 |
+
"step": 510
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 2.0,
|
| 715 |
+
"step": 512,
|
| 716 |
+
"total_flos": 4.1654970530503066e+17,
|
| 717 |
+
"train_loss": 2.223597173579037,
|
| 718 |
+
"train_runtime": 8040.0291,
|
| 719 |
+
"train_samples_per_second": 4.084,
|
| 720 |
+
"train_steps_per_second": 0.064,
|
| 721 |
+
"train_tokens_per_second": 8363.146
|
| 722 |
+
}
|
| 723 |
+
],
|
| 724 |
+
"logging_steps": 5,
|
| 725 |
+
"max_steps": 512,
|
| 726 |
+
"num_input_tokens_seen": 0,
|
| 727 |
+
"num_train_epochs": 2,
|
| 728 |
+
"save_steps": 50,
|
| 729 |
+
"total_flos": 4.1654970530503066e+17,
|
| 730 |
+
"train_batch_size": 4,
|
| 731 |
+
"trial_name": null,
|
| 732 |
+
"trial_params": null
|
| 733 |
+
}
|