Commit
•
b77f2d5
1
Parent(s):
28cb71f
Model save
Browse files- README.md +78 -0
- adapter_model.safetensors +1 -1
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +436 -0
README.md
ADDED
|
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: google/gemma-2-2b-it
|
| 3 |
+
library_name: peft
|
| 4 |
+
license: gemma
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: Gemma-2-2B_task-1_120-samples_config-2
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# Gemma-2-2B_task-1_120-samples_config-2
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [google/gemma-2-2b-it](https://huggingface.co/google/gemma-2-2b-it) on the None dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 3.1679
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 0.0001
|
| 41 |
+
- train_batch_size: 1
|
| 42 |
+
- eval_batch_size: 1
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- gradient_accumulation_steps: 16
|
| 46 |
+
- total_train_batch_size: 16
|
| 47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 48 |
+
- lr_scheduler_type: cosine
|
| 49 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 50 |
+
- num_epochs: 50
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 55 |
+
|:-------------:|:-------:|:----:|:---------------:|
|
| 56 |
+
| 2.4033 | 0.9091 | 5 | 2.3638 |
|
| 57 |
+
| 2.1644 | 2.0 | 11 | 1.9085 |
|
| 58 |
+
| 1.5867 | 2.9091 | 16 | 1.6096 |
|
| 59 |
+
| 1.3937 | 4.0 | 22 | 1.4499 |
|
| 60 |
+
| 1.2999 | 4.9091 | 27 | 1.3671 |
|
| 61 |
+
| 1.1454 | 6.0 | 33 | 1.3296 |
|
| 62 |
+
| 0.9351 | 6.9091 | 38 | 1.3268 |
|
| 63 |
+
| 0.8108 | 8.0 | 44 | 1.3552 |
|
| 64 |
+
| 0.6191 | 8.9091 | 49 | 1.4658 |
|
| 65 |
+
| 0.4271 | 10.0 | 55 | 1.7157 |
|
| 66 |
+
| 0.3163 | 10.9091 | 60 | 2.0603 |
|
| 67 |
+
| 0.1817 | 12.0 | 66 | 2.5517 |
|
| 68 |
+
| 0.104 | 12.9091 | 71 | 2.9069 |
|
| 69 |
+
| 0.0403 | 14.0 | 77 | 3.1679 |
|
| 70 |
+
|
| 71 |
+
|
| 72 |
+
### Framework versions
|
| 73 |
+
|
| 74 |
+
- PEFT 0.12.0
|
| 75 |
+
- Transformers 4.44.0
|
| 76 |
+
- Pytorch 2.1.2+cu121
|
| 77 |
+
- Datasets 2.20.0
|
| 78 |
+
- Tokenizers 0.19.1
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 83115256
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:20c29e389cde6e4bcc42f844b6221047fb61d4fe8661660c65abf7055eb31628
|
| 3 |
size 83115256
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 14.0,
|
| 3 |
+
"total_flos": 4919776673529856.0,
|
| 4 |
+
"train_loss": 0.9845529866295976,
|
| 5 |
+
"train_runtime": 544.3113,
|
| 6 |
+
"train_samples": 88,
|
| 7 |
+
"train_samples_per_second": 8.084,
|
| 8 |
+
"train_steps_per_second": 0.459
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 14.0,
|
| 3 |
+
"total_flos": 4919776673529856.0,
|
| 4 |
+
"train_loss": 0.9845529866295976,
|
| 5 |
+
"train_runtime": 544.3113,
|
| 6 |
+
"train_samples": 88,
|
| 7 |
+
"train_samples_per_second": 8.084,
|
| 8 |
+
"train_steps_per_second": 0.459
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,436 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.3267595767974854,
|
| 3 |
+
"best_model_checkpoint": "data/Gemma-2-2B_task-1_120-samples_config-2/checkpoint-38",
|
| 4 |
+
"epoch": 14.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 77,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.18181818181818182,
|
| 13 |
+
"grad_norm": 2.5245699882507324,
|
| 14 |
+
"learning_rate": 4.000000000000001e-06,
|
| 15 |
+
"loss": 2.5052,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.36363636363636365,
|
| 20 |
+
"grad_norm": 2.7616875171661377,
|
| 21 |
+
"learning_rate": 8.000000000000001e-06,
|
| 22 |
+
"loss": 2.4313,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.7272727272727273,
|
| 27 |
+
"grad_norm": 2.373145341873169,
|
| 28 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 29 |
+
"loss": 2.4033,
|
| 30 |
+
"step": 4
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.9090909090909091,
|
| 34 |
+
"eval_loss": 2.363821029663086,
|
| 35 |
+
"eval_runtime": 2.2328,
|
| 36 |
+
"eval_samples_per_second": 10.749,
|
| 37 |
+
"eval_steps_per_second": 10.749,
|
| 38 |
+
"step": 5
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.0909090909090908,
|
| 42 |
+
"grad_norm": 2.6781208515167236,
|
| 43 |
+
"learning_rate": 2.4e-05,
|
| 44 |
+
"loss": 2.3471,
|
| 45 |
+
"step": 6
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.4545454545454546,
|
| 49 |
+
"grad_norm": 2.3583812713623047,
|
| 50 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 51 |
+
"loss": 2.3318,
|
| 52 |
+
"step": 8
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.8181818181818183,
|
| 56 |
+
"grad_norm": 2.3059582710266113,
|
| 57 |
+
"learning_rate": 4e-05,
|
| 58 |
+
"loss": 2.1644,
|
| 59 |
+
"step": 10
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 2.0,
|
| 63 |
+
"eval_loss": 1.9085012674331665,
|
| 64 |
+
"eval_runtime": 2.2159,
|
| 65 |
+
"eval_samples_per_second": 10.831,
|
| 66 |
+
"eval_steps_per_second": 10.831,
|
| 67 |
+
"step": 11
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 2.1818181818181817,
|
| 71 |
+
"grad_norm": 1.878598690032959,
|
| 72 |
+
"learning_rate": 4.8e-05,
|
| 73 |
+
"loss": 1.9701,
|
| 74 |
+
"step": 12
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.5454545454545454,
|
| 78 |
+
"grad_norm": 1.6238771677017212,
|
| 79 |
+
"learning_rate": 5.6000000000000006e-05,
|
| 80 |
+
"loss": 1.7284,
|
| 81 |
+
"step": 14
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 2.909090909090909,
|
| 85 |
+
"grad_norm": 1.4010099172592163,
|
| 86 |
+
"learning_rate": 6.400000000000001e-05,
|
| 87 |
+
"loss": 1.5867,
|
| 88 |
+
"step": 16
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 2.909090909090909,
|
| 92 |
+
"eval_loss": 1.6096315383911133,
|
| 93 |
+
"eval_runtime": 2.2189,
|
| 94 |
+
"eval_samples_per_second": 10.816,
|
| 95 |
+
"eval_steps_per_second": 10.816,
|
| 96 |
+
"step": 16
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 3.2727272727272725,
|
| 100 |
+
"grad_norm": 0.9943261742591858,
|
| 101 |
+
"learning_rate": 7.2e-05,
|
| 102 |
+
"loss": 1.5703,
|
| 103 |
+
"step": 18
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 3.6363636363636362,
|
| 107 |
+
"grad_norm": 1.0258581638336182,
|
| 108 |
+
"learning_rate": 8e-05,
|
| 109 |
+
"loss": 1.4806,
|
| 110 |
+
"step": 20
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 4.0,
|
| 114 |
+
"grad_norm": 0.9659519791603088,
|
| 115 |
+
"learning_rate": 8.800000000000001e-05,
|
| 116 |
+
"loss": 1.3937,
|
| 117 |
+
"step": 22
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 4.0,
|
| 121 |
+
"eval_loss": 1.4499346017837524,
|
| 122 |
+
"eval_runtime": 2.191,
|
| 123 |
+
"eval_samples_per_second": 10.954,
|
| 124 |
+
"eval_steps_per_second": 10.954,
|
| 125 |
+
"step": 22
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 4.363636363636363,
|
| 129 |
+
"grad_norm": 0.8161854147911072,
|
| 130 |
+
"learning_rate": 9.6e-05,
|
| 131 |
+
"loss": 1.3292,
|
| 132 |
+
"step": 24
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 4.7272727272727275,
|
| 136 |
+
"grad_norm": 0.7501208782196045,
|
| 137 |
+
"learning_rate": 9.999512620046522e-05,
|
| 138 |
+
"loss": 1.2999,
|
| 139 |
+
"step": 26
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 4.909090909090909,
|
| 143 |
+
"eval_loss": 1.3671011924743652,
|
| 144 |
+
"eval_runtime": 2.2834,
|
| 145 |
+
"eval_samples_per_second": 10.511,
|
| 146 |
+
"eval_steps_per_second": 10.511,
|
| 147 |
+
"step": 27
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 5.090909090909091,
|
| 151 |
+
"grad_norm": 0.7938846349716187,
|
| 152 |
+
"learning_rate": 9.995614150494293e-05,
|
| 153 |
+
"loss": 1.2026,
|
| 154 |
+
"step": 28
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 5.454545454545454,
|
| 158 |
+
"grad_norm": 0.758736252784729,
|
| 159 |
+
"learning_rate": 9.987820251299122e-05,
|
| 160 |
+
"loss": 1.1153,
|
| 161 |
+
"step": 30
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 5.818181818181818,
|
| 165 |
+
"grad_norm": 0.8454522490501404,
|
| 166 |
+
"learning_rate": 9.976136999909156e-05,
|
| 167 |
+
"loss": 1.1454,
|
| 168 |
+
"step": 32
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 6.0,
|
| 172 |
+
"eval_loss": 1.329568862915039,
|
| 173 |
+
"eval_runtime": 2.1499,
|
| 174 |
+
"eval_samples_per_second": 11.163,
|
| 175 |
+
"eval_steps_per_second": 11.163,
|
| 176 |
+
"step": 33
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 6.181818181818182,
|
| 180 |
+
"grad_norm": 0.806365966796875,
|
| 181 |
+
"learning_rate": 9.96057350657239e-05,
|
| 182 |
+
"loss": 1.0849,
|
| 183 |
+
"step": 34
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 6.545454545454545,
|
| 187 |
+
"grad_norm": 0.8713198900222778,
|
| 188 |
+
"learning_rate": 9.941141907232765e-05,
|
| 189 |
+
"loss": 0.9951,
|
| 190 |
+
"step": 36
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 6.909090909090909,
|
| 194 |
+
"grad_norm": 0.9592146277427673,
|
| 195 |
+
"learning_rate": 9.917857354066931e-05,
|
| 196 |
+
"loss": 0.9351,
|
| 197 |
+
"step": 38
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 6.909090909090909,
|
| 201 |
+
"eval_loss": 1.3267595767974854,
|
| 202 |
+
"eval_runtime": 2.298,
|
| 203 |
+
"eval_samples_per_second": 10.444,
|
| 204 |
+
"eval_steps_per_second": 10.444,
|
| 205 |
+
"step": 38
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 7.2727272727272725,
|
| 209 |
+
"grad_norm": 0.8650667667388916,
|
| 210 |
+
"learning_rate": 9.890738003669029e-05,
|
| 211 |
+
"loss": 0.8645,
|
| 212 |
+
"step": 40
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 7.636363636363637,
|
| 216 |
+
"grad_norm": 0.9249936938285828,
|
| 217 |
+
"learning_rate": 9.859805002892732e-05,
|
| 218 |
+
"loss": 0.8049,
|
| 219 |
+
"step": 42
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 8.0,
|
| 223 |
+
"grad_norm": 1.106773853302002,
|
| 224 |
+
"learning_rate": 9.825082472361557e-05,
|
| 225 |
+
"loss": 0.8108,
|
| 226 |
+
"step": 44
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 8.0,
|
| 230 |
+
"eval_loss": 1.355177879333496,
|
| 231 |
+
"eval_runtime": 2.2733,
|
| 232 |
+
"eval_samples_per_second": 10.557,
|
| 233 |
+
"eval_steps_per_second": 10.557,
|
| 234 |
+
"step": 44
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 8.363636363636363,
|
| 238 |
+
"grad_norm": 1.0761725902557373,
|
| 239 |
+
"learning_rate": 9.786597487660337e-05,
|
| 240 |
+
"loss": 0.7095,
|
| 241 |
+
"step": 46
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 8.727272727272727,
|
| 245 |
+
"grad_norm": 1.1069787740707397,
|
| 246 |
+
"learning_rate": 9.744380058222483e-05,
|
| 247 |
+
"loss": 0.6191,
|
| 248 |
+
"step": 48
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 8.909090909090908,
|
| 252 |
+
"eval_loss": 1.4658174514770508,
|
| 253 |
+
"eval_runtime": 2.2892,
|
| 254 |
+
"eval_samples_per_second": 10.484,
|
| 255 |
+
"eval_steps_per_second": 10.484,
|
| 256 |
+
"step": 49
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 9.090909090909092,
|
| 260 |
+
"grad_norm": 1.5594687461853027,
|
| 261 |
+
"learning_rate": 9.698463103929542e-05,
|
| 262 |
+
"loss": 0.6243,
|
| 263 |
+
"step": 50
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 9.454545454545455,
|
| 267 |
+
"grad_norm": 1.2610054016113281,
|
| 268 |
+
"learning_rate": 9.648882429441257e-05,
|
| 269 |
+
"loss": 0.5089,
|
| 270 |
+
"step": 52
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 9.818181818181818,
|
| 274 |
+
"grad_norm": 1.41365647315979,
|
| 275 |
+
"learning_rate": 9.595676696276172e-05,
|
| 276 |
+
"loss": 0.4271,
|
| 277 |
+
"step": 54
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 10.0,
|
| 281 |
+
"eval_loss": 1.7157014608383179,
|
| 282 |
+
"eval_runtime": 2.2343,
|
| 283 |
+
"eval_samples_per_second": 10.741,
|
| 284 |
+
"eval_steps_per_second": 10.741,
|
| 285 |
+
"step": 55
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 10.181818181818182,
|
| 289 |
+
"grad_norm": 1.3679523468017578,
|
| 290 |
+
"learning_rate": 9.538887392664544e-05,
|
| 291 |
+
"loss": 0.4122,
|
| 292 |
+
"step": 56
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 10.545454545454545,
|
| 296 |
+
"grad_norm": 1.6108626127243042,
|
| 297 |
+
"learning_rate": 9.478558801197065e-05,
|
| 298 |
+
"loss": 0.3251,
|
| 299 |
+
"step": 58
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 10.909090909090908,
|
| 303 |
+
"grad_norm": 1.9596925973892212,
|
| 304 |
+
"learning_rate": 9.414737964294636e-05,
|
| 305 |
+
"loss": 0.3163,
|
| 306 |
+
"step": 60
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 10.909090909090908,
|
| 310 |
+
"eval_loss": 2.0603339672088623,
|
| 311 |
+
"eval_runtime": 2.2678,
|
| 312 |
+
"eval_samples_per_second": 10.583,
|
| 313 |
+
"eval_steps_per_second": 10.583,
|
| 314 |
+
"step": 60
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 11.272727272727273,
|
| 318 |
+
"grad_norm": 1.2331981658935547,
|
| 319 |
+
"learning_rate": 9.347474647526095e-05,
|
| 320 |
+
"loss": 0.1779,
|
| 321 |
+
"step": 62
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 11.636363636363637,
|
| 325 |
+
"grad_norm": 1.564605474472046,
|
| 326 |
+
"learning_rate": 9.276821300802534e-05,
|
| 327 |
+
"loss": 0.1847,
|
| 328 |
+
"step": 64
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 12.0,
|
| 332 |
+
"grad_norm": 2.0967092514038086,
|
| 333 |
+
"learning_rate": 9.202833017478422e-05,
|
| 334 |
+
"loss": 0.1817,
|
| 335 |
+
"step": 66
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 12.0,
|
| 339 |
+
"eval_loss": 2.5517404079437256,
|
| 340 |
+
"eval_runtime": 2.2226,
|
| 341 |
+
"eval_samples_per_second": 10.798,
|
| 342 |
+
"eval_steps_per_second": 10.798,
|
| 343 |
+
"step": 66
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 12.363636363636363,
|
| 347 |
+
"grad_norm": 1.0268821716308594,
|
| 348 |
+
"learning_rate": 9.125567491391476e-05,
|
| 349 |
+
"loss": 0.0953,
|
| 350 |
+
"step": 68
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 12.727272727272727,
|
| 354 |
+
"grad_norm": 1.8212908506393433,
|
| 355 |
+
"learning_rate": 9.045084971874738e-05,
|
| 356 |
+
"loss": 0.104,
|
| 357 |
+
"step": 70
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 12.909090909090908,
|
| 361 |
+
"eval_loss": 2.906930923461914,
|
| 362 |
+
"eval_runtime": 2.268,
|
| 363 |
+
"eval_samples_per_second": 10.582,
|
| 364 |
+
"eval_steps_per_second": 10.582,
|
| 365 |
+
"step": 71
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 13.090909090909092,
|
| 369 |
+
"grad_norm": 1.7785996198654175,
|
| 370 |
+
"learning_rate": 8.961448216775954e-05,
|
| 371 |
+
"loss": 0.076,
|
| 372 |
+
"step": 72
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 13.454545454545455,
|
| 376 |
+
"grad_norm": 1.1120202541351318,
|
| 377 |
+
"learning_rate": 8.874722443520899e-05,
|
| 378 |
+
"loss": 0.0483,
|
| 379 |
+
"step": 74
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 13.818181818181818,
|
| 383 |
+
"grad_norm": 1.672891616821289,
|
| 384 |
+
"learning_rate": 8.784975278258783e-05,
|
| 385 |
+
"loss": 0.0403,
|
| 386 |
+
"step": 76
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 14.0,
|
| 390 |
+
"eval_loss": 3.1679420471191406,
|
| 391 |
+
"eval_runtime": 2.2313,
|
| 392 |
+
"eval_samples_per_second": 10.756,
|
| 393 |
+
"eval_steps_per_second": 10.756,
|
| 394 |
+
"step": 77
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 14.0,
|
| 398 |
+
"step": 77,
|
| 399 |
+
"total_flos": 4919776673529856.0,
|
| 400 |
+
"train_loss": 0.9845529866295976,
|
| 401 |
+
"train_runtime": 544.3113,
|
| 402 |
+
"train_samples_per_second": 8.084,
|
| 403 |
+
"train_steps_per_second": 0.459
|
| 404 |
+
}
|
| 405 |
+
],
|
| 406 |
+
"logging_steps": 2,
|
| 407 |
+
"max_steps": 250,
|
| 408 |
+
"num_input_tokens_seen": 0,
|
| 409 |
+
"num_train_epochs": 50,
|
| 410 |
+
"save_steps": 25,
|
| 411 |
+
"stateful_callbacks": {
|
| 412 |
+
"EarlyStoppingCallback": {
|
| 413 |
+
"args": {
|
| 414 |
+
"early_stopping_patience": 7,
|
| 415 |
+
"early_stopping_threshold": 0.0
|
| 416 |
+
},
|
| 417 |
+
"attributes": {
|
| 418 |
+
"early_stopping_patience_counter": 0
|
| 419 |
+
}
|
| 420 |
+
},
|
| 421 |
+
"TrainerControl": {
|
| 422 |
+
"args": {
|
| 423 |
+
"should_epoch_stop": false,
|
| 424 |
+
"should_evaluate": false,
|
| 425 |
+
"should_log": false,
|
| 426 |
+
"should_save": true,
|
| 427 |
+
"should_training_stop": true
|
| 428 |
+
},
|
| 429 |
+
"attributes": {}
|
| 430 |
+
}
|
| 431 |
+
},
|
| 432 |
+
"total_flos": 4919776673529856.0,
|
| 433 |
+
"train_batch_size": 1,
|
| 434 |
+
"trial_name": null,
|
| 435 |
+
"trial_params": null
|
| 436 |
+
}
|