Model save
Browse files- README.md +65 -0
- all_results.json +9 -0
- generation_config.json +7 -0
- model.safetensors +1 -1
- runs/Jul12_10-47-58_poseidon/events.out.tfevents.1720781840.poseidon.3256517.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +568 -0
README.md
ADDED
|
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
datasets:
|
| 9 |
+
- generator
|
| 10 |
+
model-index:
|
| 11 |
+
- name: tinyllama-1.1b-mt-sft-full_new
|
| 12 |
+
results: []
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
+
|
| 18 |
+
# tinyllama-1.1b-mt-sft-full_new
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T](https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T) on the generator dataset.
|
| 21 |
+
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 1.7292
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 1e-05
|
| 42 |
+
- train_batch_size: 4
|
| 43 |
+
- eval_batch_size: 8
|
| 44 |
+
- seed: 42
|
| 45 |
+
- distributed_type: multi-GPU
|
| 46 |
+
- gradient_accumulation_steps: 2
|
| 47 |
+
- total_train_batch_size: 8
|
| 48 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 49 |
+
- lr_scheduler_type: cosine
|
| 50 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 51 |
+
- num_epochs: 1
|
| 52 |
+
|
| 53 |
+
### Training results
|
| 54 |
+
|
| 55 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 56 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 57 |
+
| 1.6671 | 1.0 | 739 | 1.7292 |
|
| 58 |
+
|
| 59 |
+
|
| 60 |
+
### Framework versions
|
| 61 |
+
|
| 62 |
+
- Transformers 4.41.2
|
| 63 |
+
- Pytorch 2.1.2
|
| 64 |
+
- Datasets 2.20.0
|
| 65 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 7.51411531916247e+16,
|
| 4 |
+
"train_loss": 1.726003464891075,
|
| 5 |
+
"train_runtime": 1754.8001,
|
| 6 |
+
"train_samples": 117404,
|
| 7 |
+
"train_samples_per_second": 3.368,
|
| 8 |
+
"train_steps_per_second": 0.421
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 1,
|
| 3 |
+
"eos_token_id": 2,
|
| 4 |
+
"max_length": 2048,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.41.2"
|
| 7 |
+
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2200119864
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e706fe2a905adcb81631b2bd4fa09de4d318d8d02944b9c5a5564e245a8462a6
|
| 3 |
size 2200119864
|
runs/Jul12_10-47-58_poseidon/events.out.tfevents.1720781840.poseidon.3256517.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8fafe339938aac22eaf4cdaac081723872de0fe3f44f51955170c2f7ebeaf31c
|
| 3 |
+
size 21231
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 7.51411531916247e+16,
|
| 4 |
+
"train_loss": 1.726003464891075,
|
| 5 |
+
"train_runtime": 1754.8001,
|
| 6 |
+
"train_samples": 117404,
|
| 7 |
+
"train_samples_per_second": 3.368,
|
| 8 |
+
"train_steps_per_second": 0.421
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,568 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 739,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0013531799729364006,
|
| 13 |
+
"grad_norm": 22.625,
|
| 14 |
+
"learning_rate": 1.3513513513513515e-07,
|
| 15 |
+
"loss": 2.1989,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.013531799729364006,
|
| 20 |
+
"grad_norm": 22.5,
|
| 21 |
+
"learning_rate": 1.3513513513513515e-06,
|
| 22 |
+
"loss": 2.2213,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.02706359945872801,
|
| 27 |
+
"grad_norm": 8.6875,
|
| 28 |
+
"learning_rate": 2.702702702702703e-06,
|
| 29 |
+
"loss": 2.2125,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.04059539918809202,
|
| 34 |
+
"grad_norm": 5.15625,
|
| 35 |
+
"learning_rate": 4.0540540540540545e-06,
|
| 36 |
+
"loss": 2.1298,
|
| 37 |
+
"step": 30
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.05412719891745602,
|
| 41 |
+
"grad_norm": 3.734375,
|
| 42 |
+
"learning_rate": 5.405405405405406e-06,
|
| 43 |
+
"loss": 2.0609,
|
| 44 |
+
"step": 40
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.06765899864682003,
|
| 48 |
+
"grad_norm": 3.65625,
|
| 49 |
+
"learning_rate": 6.7567567567567575e-06,
|
| 50 |
+
"loss": 2.0108,
|
| 51 |
+
"step": 50
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.08119079837618404,
|
| 55 |
+
"grad_norm": 3.546875,
|
| 56 |
+
"learning_rate": 8.108108108108109e-06,
|
| 57 |
+
"loss": 1.918,
|
| 58 |
+
"step": 60
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.09472259810554803,
|
| 62 |
+
"grad_norm": 3.03125,
|
| 63 |
+
"learning_rate": 9.45945945945946e-06,
|
| 64 |
+
"loss": 1.8292,
|
| 65 |
+
"step": 70
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.10825439783491204,
|
| 69 |
+
"grad_norm": 2.90625,
|
| 70 |
+
"learning_rate": 9.99799150901764e-06,
|
| 71 |
+
"loss": 1.8037,
|
| 72 |
+
"step": 80
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.12178619756427606,
|
| 76 |
+
"grad_norm": 2.796875,
|
| 77 |
+
"learning_rate": 9.985723240513795e-06,
|
| 78 |
+
"loss": 1.7704,
|
| 79 |
+
"step": 90
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.13531799729364005,
|
| 83 |
+
"grad_norm": 2.578125,
|
| 84 |
+
"learning_rate": 9.962329873928743e-06,
|
| 85 |
+
"loss": 1.7527,
|
| 86 |
+
"step": 100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.14884979702300405,
|
| 90 |
+
"grad_norm": 2.609375,
|
| 91 |
+
"learning_rate": 9.92786360901067e-06,
|
| 92 |
+
"loss": 1.7479,
|
| 93 |
+
"step": 110
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.16238159675236807,
|
| 97 |
+
"grad_norm": 2.640625,
|
| 98 |
+
"learning_rate": 9.882401353473711e-06,
|
| 99 |
+
"loss": 1.7169,
|
| 100 |
+
"step": 120
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.17591339648173207,
|
| 104 |
+
"grad_norm": 2.625,
|
| 105 |
+
"learning_rate": 9.826044551386743e-06,
|
| 106 |
+
"loss": 1.7196,
|
| 107 |
+
"step": 130
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.18944519621109607,
|
| 111 |
+
"grad_norm": 2.703125,
|
| 112 |
+
"learning_rate": 9.758918956812024e-06,
|
| 113 |
+
"loss": 1.7238,
|
| 114 |
+
"step": 140
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.2029769959404601,
|
| 118 |
+
"grad_norm": 2.65625,
|
| 119 |
+
"learning_rate": 9.681174353198687e-06,
|
| 120 |
+
"loss": 1.7291,
|
| 121 |
+
"step": 150
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.2165087956698241,
|
| 125 |
+
"grad_norm": 2.625,
|
| 126 |
+
"learning_rate": 9.59298421915731e-06,
|
| 127 |
+
"loss": 1.7071,
|
| 128 |
+
"step": 160
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.23004059539918809,
|
| 132 |
+
"grad_norm": 2.546875,
|
| 133 |
+
"learning_rate": 9.494545341361291e-06,
|
| 134 |
+
"loss": 1.715,
|
| 135 |
+
"step": 170
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.2435723951285521,
|
| 139 |
+
"grad_norm": 2.734375,
|
| 140 |
+
"learning_rate": 9.386077375438848e-06,
|
| 141 |
+
"loss": 1.7119,
|
| 142 |
+
"step": 180
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.2571041948579161,
|
| 146 |
+
"grad_norm": 2.5625,
|
| 147 |
+
"learning_rate": 9.267822355835402e-06,
|
| 148 |
+
"loss": 1.7079,
|
| 149 |
+
"step": 190
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.2706359945872801,
|
| 153 |
+
"grad_norm": 2.6875,
|
| 154 |
+
"learning_rate": 9.140044155740102e-06,
|
| 155 |
+
"loss": 1.6912,
|
| 156 |
+
"step": 200
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.28416779431664413,
|
| 160 |
+
"grad_norm": 2.6875,
|
| 161 |
+
"learning_rate": 9.003027898281551e-06,
|
| 162 |
+
"loss": 1.7097,
|
| 163 |
+
"step": 210
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.2976995940460081,
|
| 167 |
+
"grad_norm": 2.59375,
|
| 168 |
+
"learning_rate": 8.85707932030663e-06,
|
| 169 |
+
"loss": 1.6883,
|
| 170 |
+
"step": 220
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.3112313937753721,
|
| 174 |
+
"grad_norm": 2.484375,
|
| 175 |
+
"learning_rate": 8.702524090162023e-06,
|
| 176 |
+
"loss": 1.6917,
|
| 177 |
+
"step": 230
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.32476319350473615,
|
| 181 |
+
"grad_norm": 2.65625,
|
| 182 |
+
"learning_rate": 8.539707081000808e-06,
|
| 183 |
+
"loss": 1.6951,
|
| 184 |
+
"step": 240
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.3382949932341001,
|
| 188 |
+
"grad_norm": 2.46875,
|
| 189 |
+
"learning_rate": 8.36899160123559e-06,
|
| 190 |
+
"loss": 1.7043,
|
| 191 |
+
"step": 250
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.35182679296346414,
|
| 195 |
+
"grad_norm": 2.5,
|
| 196 |
+
"learning_rate": 8.190758583855379e-06,
|
| 197 |
+
"loss": 1.6851,
|
| 198 |
+
"step": 260
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.36535859269282817,
|
| 202 |
+
"grad_norm": 2.53125,
|
| 203 |
+
"learning_rate": 8.005405736415127e-06,
|
| 204 |
+
"loss": 1.6912,
|
| 205 |
+
"step": 270
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.37889039242219213,
|
| 209 |
+
"grad_norm": 2.578125,
|
| 210 |
+
"learning_rate": 7.813346653594667e-06,
|
| 211 |
+
"loss": 1.7011,
|
| 212 |
+
"step": 280
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.39242219215155616,
|
| 216 |
+
"grad_norm": 2.484375,
|
| 217 |
+
"learning_rate": 7.615009894307263e-06,
|
| 218 |
+
"loss": 1.6955,
|
| 219 |
+
"step": 290
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.4059539918809202,
|
| 223 |
+
"grad_norm": 2.640625,
|
| 224 |
+
"learning_rate": 7.410838025417083e-06,
|
| 225 |
+
"loss": 1.6848,
|
| 226 |
+
"step": 300
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.41948579161028415,
|
| 230 |
+
"grad_norm": 2.6875,
|
| 231 |
+
"learning_rate": 7.201286634199484e-06,
|
| 232 |
+
"loss": 1.6749,
|
| 233 |
+
"step": 310
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.4330175913396482,
|
| 237 |
+
"grad_norm": 2.546875,
|
| 238 |
+
"learning_rate": 6.986823311747652e-06,
|
| 239 |
+
"loss": 1.6797,
|
| 240 |
+
"step": 320
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.4465493910690122,
|
| 244 |
+
"grad_norm": 2.53125,
|
| 245 |
+
"learning_rate": 6.767926609594032e-06,
|
| 246 |
+
"loss": 1.6744,
|
| 247 |
+
"step": 330
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.46008119079837617,
|
| 251 |
+
"grad_norm": 2.453125,
|
| 252 |
+
"learning_rate": 6.545084971874738e-06,
|
| 253 |
+
"loss": 1.6636,
|
| 254 |
+
"step": 340
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.4736129905277402,
|
| 258 |
+
"grad_norm": 2.640625,
|
| 259 |
+
"learning_rate": 6.3187956454196885e-06,
|
| 260 |
+
"loss": 1.6949,
|
| 261 |
+
"step": 350
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.4871447902571042,
|
| 265 |
+
"grad_norm": 2.46875,
|
| 266 |
+
"learning_rate": 6.0895635702004985e-06,
|
| 267 |
+
"loss": 1.6903,
|
| 268 |
+
"step": 360
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.5006765899864682,
|
| 272 |
+
"grad_norm": 2.53125,
|
| 273 |
+
"learning_rate": 5.857900252611959e-06,
|
| 274 |
+
"loss": 1.6602,
|
| 275 |
+
"step": 370
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.5142083897158322,
|
| 279 |
+
"grad_norm": 2.515625,
|
| 280 |
+
"learning_rate": 5.624322624101255e-06,
|
| 281 |
+
"loss": 1.6888,
|
| 282 |
+
"step": 380
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.5277401894451962,
|
| 286 |
+
"grad_norm": 2.765625,
|
| 287 |
+
"learning_rate": 5.3893518876917795e-06,
|
| 288 |
+
"loss": 1.6876,
|
| 289 |
+
"step": 390
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.5412719891745602,
|
| 293 |
+
"grad_norm": 2.546875,
|
| 294 |
+
"learning_rate": 5.153512354975388e-06,
|
| 295 |
+
"loss": 1.656,
|
| 296 |
+
"step": 400
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.5548037889039242,
|
| 300 |
+
"grad_norm": 2.71875,
|
| 301 |
+
"learning_rate": 4.917330276168208e-06,
|
| 302 |
+
"loss": 1.6737,
|
| 303 |
+
"step": 410
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.5683355886332883,
|
| 307 |
+
"grad_norm": 2.59375,
|
| 308 |
+
"learning_rate": 4.681332665840647e-06,
|
| 309 |
+
"loss": 1.6856,
|
| 310 |
+
"step": 420
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.5818673883626523,
|
| 314 |
+
"grad_norm": 2.5625,
|
| 315 |
+
"learning_rate": 4.446046126941801e-06,
|
| 316 |
+
"loss": 1.6917,
|
| 317 |
+
"step": 430
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.5953991880920162,
|
| 321 |
+
"grad_norm": 2.71875,
|
| 322 |
+
"learning_rate": 4.211995675742358e-06,
|
| 323 |
+
"loss": 1.6728,
|
| 324 |
+
"step": 440
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.6089309878213802,
|
| 328 |
+
"grad_norm": 2.59375,
|
| 329 |
+
"learning_rate": 3.979703570318017e-06,
|
| 330 |
+
"loss": 1.6919,
|
| 331 |
+
"step": 450
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.6224627875507442,
|
| 335 |
+
"grad_norm": 2.609375,
|
| 336 |
+
"learning_rate": 3.749688145187497e-06,
|
| 337 |
+
"loss": 1.6789,
|
| 338 |
+
"step": 460
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.6359945872801083,
|
| 342 |
+
"grad_norm": 2.578125,
|
| 343 |
+
"learning_rate": 3.5224626547055463e-06,
|
| 344 |
+
"loss": 1.687,
|
| 345 |
+
"step": 470
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.6495263870094723,
|
| 349 |
+
"grad_norm": 2.703125,
|
| 350 |
+
"learning_rate": 3.298534127791785e-06,
|
| 351 |
+
"loss": 1.678,
|
| 352 |
+
"step": 480
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.6630581867388363,
|
| 356 |
+
"grad_norm": 2.453125,
|
| 357 |
+
"learning_rate": 3.078402236550926e-06,
|
| 358 |
+
"loss": 1.6756,
|
| 359 |
+
"step": 490
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.6765899864682002,
|
| 363 |
+
"grad_norm": 2.546875,
|
| 364 |
+
"learning_rate": 2.86255818130892e-06,
|
| 365 |
+
"loss": 1.6895,
|
| 366 |
+
"step": 500
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.6901217861975643,
|
| 370 |
+
"grad_norm": 2.5,
|
| 371 |
+
"learning_rate": 2.6514835945529706e-06,
|
| 372 |
+
"loss": 1.685,
|
| 373 |
+
"step": 510
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.7036535859269283,
|
| 377 |
+
"grad_norm": 2.609375,
|
| 378 |
+
"learning_rate": 2.4456494662211082e-06,
|
| 379 |
+
"loss": 1.6879,
|
| 380 |
+
"step": 520
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.7171853856562923,
|
| 384 |
+
"grad_norm": 2.59375,
|
| 385 |
+
"learning_rate": 2.245515092739488e-06,
|
| 386 |
+
"loss": 1.6812,
|
| 387 |
+
"step": 530
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.7307171853856563,
|
| 391 |
+
"grad_norm": 2.671875,
|
| 392 |
+
"learning_rate": 2.0515270521524562e-06,
|
| 393 |
+
"loss": 1.6727,
|
| 394 |
+
"step": 540
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.7442489851150202,
|
| 398 |
+
"grad_norm": 2.5625,
|
| 399 |
+
"learning_rate": 1.864118207632315e-06,
|
| 400 |
+
"loss": 1.6839,
|
| 401 |
+
"step": 550
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.7577807848443843,
|
| 405 |
+
"grad_norm": 2.78125,
|
| 406 |
+
"learning_rate": 1.683706741592327e-06,
|
| 407 |
+
"loss": 1.6745,
|
| 408 |
+
"step": 560
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.7713125845737483,
|
| 412 |
+
"grad_norm": 2.796875,
|
| 413 |
+
"learning_rate": 1.5106952225582312e-06,
|
| 414 |
+
"loss": 1.677,
|
| 415 |
+
"step": 570
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.7848443843031123,
|
| 419 |
+
"grad_norm": 2.53125,
|
| 420 |
+
"learning_rate": 1.3454697068804434e-06,
|
| 421 |
+
"loss": 1.6677,
|
| 422 |
+
"step": 580
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.7983761840324763,
|
| 426 |
+
"grad_norm": 2.515625,
|
| 427 |
+
"learning_rate": 1.1883988772913924e-06,
|
| 428 |
+
"loss": 1.6725,
|
| 429 |
+
"step": 590
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.8119079837618404,
|
| 433 |
+
"grad_norm": 2.5625,
|
| 434 |
+
"learning_rate": 1.0398332202301708e-06,
|
| 435 |
+
"loss": 1.669,
|
| 436 |
+
"step": 600
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.8254397834912043,
|
| 440 |
+
"grad_norm": 2.515625,
|
| 441 |
+
"learning_rate": 9.001042437702468e-07,
|
| 442 |
+
"loss": 1.6775,
|
| 443 |
+
"step": 610
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.8389715832205683,
|
| 447 |
+
"grad_norm": 2.578125,
|
| 448 |
+
"learning_rate": 7.695237378953224e-07,
|
| 449 |
+
"loss": 1.6914,
|
| 450 |
+
"step": 620
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.8525033829499323,
|
| 454 |
+
"grad_norm": 2.5,
|
| 455 |
+
"learning_rate": 6.483830787739659e-07,
|
| 456 |
+
"loss": 1.6743,
|
| 457 |
+
"step": 630
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.8660351826792964,
|
| 461 |
+
"grad_norm": 2.453125,
|
| 462 |
+
"learning_rate": 5.369525785854368e-07,
|
| 463 |
+
"loss": 1.6784,
|
| 464 |
+
"step": 640
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.8795669824086604,
|
| 468 |
+
"grad_norm": 2.546875,
|
| 469 |
+
"learning_rate": 4.3548088234752814e-07,
|
| 470 |
+
"loss": 1.687,
|
| 471 |
+
"step": 650
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.8930987821380244,
|
| 475 |
+
"grad_norm": 2.59375,
|
| 476 |
+
"learning_rate": 3.4419441309229587e-07,
|
| 477 |
+
"loss": 1.6729,
|
| 478 |
+
"step": 660
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.9066305818673883,
|
| 482 |
+
"grad_norm": 2.5625,
|
| 483 |
+
"learning_rate": 2.6329686662774247e-07,
|
| 484 |
+
"loss": 1.6775,
|
| 485 |
+
"step": 670
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.9201623815967523,
|
| 489 |
+
"grad_norm": 2.390625,
|
| 490 |
+
"learning_rate": 1.9296875701281858e-07,
|
| 491 |
+
"loss": 1.6813,
|
| 492 |
+
"step": 680
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.9336941813261164,
|
| 496 |
+
"grad_norm": 2.46875,
|
| 497 |
+
"learning_rate": 1.333670137599713e-07,
|
| 498 |
+
"loss": 1.6944,
|
| 499 |
+
"step": 690
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.9472259810554804,
|
| 503 |
+
"grad_norm": 2.75,
|
| 504 |
+
"learning_rate": 8.462463166403978e-08,
|
| 505 |
+
"loss": 1.6899,
|
| 506 |
+
"step": 700
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.9607577807848444,
|
| 510 |
+
"grad_norm": 2.4375,
|
| 511 |
+
"learning_rate": 4.685037403886483e-08,
|
| 512 |
+
"loss": 1.6669,
|
| 513 |
+
"step": 710
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.9742895805142084,
|
| 517 |
+
"grad_norm": 2.640625,
|
| 518 |
+
"learning_rate": 2.012853002380466e-08,
|
| 519 |
+
"loss": 1.6881,
|
| 520 |
+
"step": 720
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.9878213802435724,
|
| 524 |
+
"grad_norm": 2.59375,
|
| 525 |
+
"learning_rate": 4.51872650170937e-09,
|
| 526 |
+
"loss": 1.6671,
|
| 527 |
+
"step": 730
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.0,
|
| 531 |
+
"eval_loss": 1.7291793823242188,
|
| 532 |
+
"eval_runtime": 27.3164,
|
| 533 |
+
"eval_samples_per_second": 15.924,
|
| 534 |
+
"eval_steps_per_second": 2.013,
|
| 535 |
+
"step": 739
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 1.0,
|
| 539 |
+
"step": 739,
|
| 540 |
+
"total_flos": 7.51411531916247e+16,
|
| 541 |
+
"train_loss": 1.726003464891075,
|
| 542 |
+
"train_runtime": 1754.8001,
|
| 543 |
+
"train_samples_per_second": 3.368,
|
| 544 |
+
"train_steps_per_second": 0.421
|
| 545 |
+
}
|
| 546 |
+
],
|
| 547 |
+
"logging_steps": 10,
|
| 548 |
+
"max_steps": 739,
|
| 549 |
+
"num_input_tokens_seen": 0,
|
| 550 |
+
"num_train_epochs": 1,
|
| 551 |
+
"save_steps": 100,
|
| 552 |
+
"stateful_callbacks": {
|
| 553 |
+
"TrainerControl": {
|
| 554 |
+
"args": {
|
| 555 |
+
"should_epoch_stop": false,
|
| 556 |
+
"should_evaluate": false,
|
| 557 |
+
"should_log": false,
|
| 558 |
+
"should_save": true,
|
| 559 |
+
"should_training_stop": false
|
| 560 |
+
},
|
| 561 |
+
"attributes": {}
|
| 562 |
+
}
|
| 563 |
+
},
|
| 564 |
+
"total_flos": 7.51411531916247e+16,
|
| 565 |
+
"train_batch_size": 4,
|
| 566 |
+
"trial_name": null,
|
| 567 |
+
"trial_params": null
|
| 568 |
+
}
|