Model save
Browse files
README.md
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
datasets:
|
| 9 |
+
- generator
|
| 10 |
+
base_model: microsoft/phi-2
|
| 11 |
+
model-index:
|
| 12 |
+
- name: phi-2-sft-ultrachat-qlora
|
| 13 |
+
results: []
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
# phi-2-sft-ultrachat-qlora
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) on the generator dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 1.2498
|
| 24 |
+
|
| 25 |
+
## Model description
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Intended uses & limitations
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training and evaluation data
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Training procedure
|
| 38 |
+
|
| 39 |
+
### Training hyperparameters
|
| 40 |
+
|
| 41 |
+
The following hyperparameters were used during training:
|
| 42 |
+
- learning_rate: 0.0002
|
| 43 |
+
- train_batch_size: 4
|
| 44 |
+
- eval_batch_size: 4
|
| 45 |
+
- seed: 42
|
| 46 |
+
- distributed_type: multi-GPU
|
| 47 |
+
- num_devices: 4
|
| 48 |
+
- gradient_accumulation_steps: 2
|
| 49 |
+
- total_train_batch_size: 32
|
| 50 |
+
- total_eval_batch_size: 16
|
| 51 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 52 |
+
- lr_scheduler_type: cosine
|
| 53 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 54 |
+
- training_steps: 500
|
| 55 |
+
|
| 56 |
+
### Training results
|
| 57 |
+
|
| 58 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 60 |
+
| 1.2922 | 0.06 | 500 | 1.2498 |
|
| 61 |
+
|
| 62 |
+
|
| 63 |
+
### Framework versions
|
| 64 |
+
|
| 65 |
+
- PEFT 0.7.1
|
| 66 |
+
- Transformers 4.36.2
|
| 67 |
+
- Pytorch 2.1.2+cu118
|
| 68 |
+
- Datasets 2.14.6
|
| 69 |
+
- Tokenizers 0.15.2
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.06,
|
| 3 |
+
"eval_loss": 1.249782919883728,
|
| 4 |
+
"eval_runtime": 1715.5628,
|
| 5 |
+
"eval_samples": 23110,
|
| 6 |
+
"eval_samples_per_second": 16.5,
|
| 7 |
+
"eval_steps_per_second": 1.032,
|
| 8 |
+
"train_loss": 1.3271520497798919,
|
| 9 |
+
"train_runtime": 4460.9425,
|
| 10 |
+
"train_samples": 207865,
|
| 11 |
+
"train_samples_per_second": 3.587,
|
| 12 |
+
"train_steps_per_second": 0.112
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.06,
|
| 3 |
+
"eval_loss": 1.249782919883728,
|
| 4 |
+
"eval_runtime": 1715.5628,
|
| 5 |
+
"eval_samples": 23110,
|
| 6 |
+
"eval_samples_per_second": 16.5,
|
| 7 |
+
"eval_steps_per_second": 1.032
|
| 8 |
+
}
|
runs/Feb28_16-26-17_gpu4-119-4/events.out.tfevents.1709098026.gpu4-119-4.9226.0
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:a68d474e5c94baf3f39951d3f102f4756423475328661c74293501ddbc18315b
|
| 3 |
+
size 21127
|
runs/Feb28_16-26-17_gpu4-119-4/events.out.tfevents.1709104203.gpu4-119-4.9226.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fad4243b6e858d3de0b1a0993c246aa41a282a4c480775334eafee14361bedcc
|
| 3 |
+
size 359
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.06,
|
| 3 |
+
"train_loss": 1.3271520497798919,
|
| 4 |
+
"train_runtime": 4460.9425,
|
| 5 |
+
"train_samples": 207865,
|
| 6 |
+
"train_samples_per_second": 3.587,
|
| 7 |
+
"train_steps_per_second": 0.112
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,644 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.06255082254331644,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0,
|
| 13 |
+
"learning_rate": 4.000000000000001e-06,
|
| 14 |
+
"loss": 1.8341,
|
| 15 |
+
"step": 1
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.0,
|
| 19 |
+
"learning_rate": 2e-05,
|
| 20 |
+
"loss": 1.9527,
|
| 21 |
+
"step": 5
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.0,
|
| 25 |
+
"learning_rate": 4e-05,
|
| 26 |
+
"loss": 1.9158,
|
| 27 |
+
"step": 10
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.0,
|
| 31 |
+
"learning_rate": 6e-05,
|
| 32 |
+
"loss": 1.8797,
|
| 33 |
+
"step": 15
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.0,
|
| 37 |
+
"learning_rate": 8e-05,
|
| 38 |
+
"loss": 1.7949,
|
| 39 |
+
"step": 20
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.0,
|
| 43 |
+
"learning_rate": 0.0001,
|
| 44 |
+
"loss": 1.7219,
|
| 45 |
+
"step": 25
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.0,
|
| 49 |
+
"learning_rate": 0.00012,
|
| 50 |
+
"loss": 1.623,
|
| 51 |
+
"step": 30
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.0,
|
| 55 |
+
"learning_rate": 0.00014,
|
| 56 |
+
"loss": 1.5828,
|
| 57 |
+
"step": 35
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.01,
|
| 61 |
+
"learning_rate": 0.00016,
|
| 62 |
+
"loss": 1.5208,
|
| 63 |
+
"step": 40
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.01,
|
| 67 |
+
"learning_rate": 0.00018,
|
| 68 |
+
"loss": 1.4697,
|
| 69 |
+
"step": 45
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.01,
|
| 73 |
+
"learning_rate": 0.0002,
|
| 74 |
+
"loss": 1.4292,
|
| 75 |
+
"step": 50
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.01,
|
| 79 |
+
"learning_rate": 0.0001999390827019096,
|
| 80 |
+
"loss": 1.4127,
|
| 81 |
+
"step": 55
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.01,
|
| 85 |
+
"learning_rate": 0.00019975640502598244,
|
| 86 |
+
"loss": 1.3752,
|
| 87 |
+
"step": 60
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.01,
|
| 91 |
+
"learning_rate": 0.00019945218953682734,
|
| 92 |
+
"loss": 1.338,
|
| 93 |
+
"step": 65
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.01,
|
| 97 |
+
"learning_rate": 0.00019902680687415705,
|
| 98 |
+
"loss": 1.325,
|
| 99 |
+
"step": 70
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.01,
|
| 103 |
+
"learning_rate": 0.00019848077530122083,
|
| 104 |
+
"loss": 1.3627,
|
| 105 |
+
"step": 75
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.01,
|
| 109 |
+
"learning_rate": 0.00019781476007338058,
|
| 110 |
+
"loss": 1.3026,
|
| 111 |
+
"step": 80
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.01,
|
| 115 |
+
"learning_rate": 0.00019702957262759965,
|
| 116 |
+
"loss": 1.3302,
|
| 117 |
+
"step": 85
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.01,
|
| 121 |
+
"learning_rate": 0.0001961261695938319,
|
| 122 |
+
"loss": 1.3032,
|
| 123 |
+
"step": 90
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.01,
|
| 127 |
+
"learning_rate": 0.00019510565162951537,
|
| 128 |
+
"loss": 1.3538,
|
| 129 |
+
"step": 95
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.01,
|
| 133 |
+
"learning_rate": 0.00019396926207859084,
|
| 134 |
+
"loss": 1.3524,
|
| 135 |
+
"step": 100
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.01,
|
| 139 |
+
"learning_rate": 0.00019271838545667876,
|
| 140 |
+
"loss": 1.2811,
|
| 141 |
+
"step": 105
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.01,
|
| 145 |
+
"learning_rate": 0.0001913545457642601,
|
| 146 |
+
"loss": 1.2991,
|
| 147 |
+
"step": 110
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.01,
|
| 151 |
+
"learning_rate": 0.0001898794046299167,
|
| 152 |
+
"loss": 1.314,
|
| 153 |
+
"step": 115
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.02,
|
| 157 |
+
"learning_rate": 0.00018829475928589271,
|
| 158 |
+
"loss": 1.2834,
|
| 159 |
+
"step": 120
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.02,
|
| 163 |
+
"learning_rate": 0.00018660254037844388,
|
| 164 |
+
"loss": 1.3266,
|
| 165 |
+
"step": 125
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.02,
|
| 169 |
+
"learning_rate": 0.0001848048096156426,
|
| 170 |
+
"loss": 1.294,
|
| 171 |
+
"step": 130
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.02,
|
| 175 |
+
"learning_rate": 0.00018290375725550417,
|
| 176 |
+
"loss": 1.3133,
|
| 177 |
+
"step": 135
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.02,
|
| 181 |
+
"learning_rate": 0.00018090169943749476,
|
| 182 |
+
"loss": 1.3191,
|
| 183 |
+
"step": 140
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 0.02,
|
| 187 |
+
"learning_rate": 0.00017880107536067218,
|
| 188 |
+
"loss": 1.2459,
|
| 189 |
+
"step": 145
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.02,
|
| 193 |
+
"learning_rate": 0.0001766044443118978,
|
| 194 |
+
"loss": 1.3149,
|
| 195 |
+
"step": 150
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.02,
|
| 199 |
+
"learning_rate": 0.00017431448254773944,
|
| 200 |
+
"loss": 1.2173,
|
| 201 |
+
"step": 155
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.02,
|
| 205 |
+
"learning_rate": 0.0001719339800338651,
|
| 206 |
+
"loss": 1.277,
|
| 207 |
+
"step": 160
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.02,
|
| 211 |
+
"learning_rate": 0.00016946583704589973,
|
| 212 |
+
"loss": 1.2873,
|
| 213 |
+
"step": 165
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.02,
|
| 217 |
+
"learning_rate": 0.00016691306063588583,
|
| 218 |
+
"loss": 1.2535,
|
| 219 |
+
"step": 170
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.02,
|
| 223 |
+
"learning_rate": 0.00016427876096865394,
|
| 224 |
+
"loss": 1.2991,
|
| 225 |
+
"step": 175
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.02,
|
| 229 |
+
"learning_rate": 0.0001615661475325658,
|
| 230 |
+
"loss": 1.2486,
|
| 231 |
+
"step": 180
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 0.02,
|
| 235 |
+
"learning_rate": 0.00015877852522924732,
|
| 236 |
+
"loss": 1.3577,
|
| 237 |
+
"step": 185
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.02,
|
| 241 |
+
"learning_rate": 0.0001559192903470747,
|
| 242 |
+
"loss": 1.2981,
|
| 243 |
+
"step": 190
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.02,
|
| 247 |
+
"learning_rate": 0.0001529919264233205,
|
| 248 |
+
"loss": 1.3271,
|
| 249 |
+
"step": 195
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.03,
|
| 253 |
+
"learning_rate": 0.00015000000000000001,
|
| 254 |
+
"loss": 1.3087,
|
| 255 |
+
"step": 200
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.03,
|
| 259 |
+
"learning_rate": 0.00014694715627858908,
|
| 260 |
+
"loss": 1.2858,
|
| 261 |
+
"step": 205
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.03,
|
| 265 |
+
"learning_rate": 0.00014383711467890774,
|
| 266 |
+
"loss": 1.306,
|
| 267 |
+
"step": 210
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.03,
|
| 271 |
+
"learning_rate": 0.00014067366430758004,
|
| 272 |
+
"loss": 1.279,
|
| 273 |
+
"step": 215
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.03,
|
| 277 |
+
"learning_rate": 0.00013746065934159123,
|
| 278 |
+
"loss": 1.3175,
|
| 279 |
+
"step": 220
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.03,
|
| 283 |
+
"learning_rate": 0.00013420201433256689,
|
| 284 |
+
"loss": 1.2362,
|
| 285 |
+
"step": 225
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.03,
|
| 289 |
+
"learning_rate": 0.00013090169943749476,
|
| 290 |
+
"loss": 1.2573,
|
| 291 |
+
"step": 230
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 0.03,
|
| 295 |
+
"learning_rate": 0.0001275637355816999,
|
| 296 |
+
"loss": 1.2666,
|
| 297 |
+
"step": 235
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.03,
|
| 301 |
+
"learning_rate": 0.00012419218955996676,
|
| 302 |
+
"loss": 1.2854,
|
| 303 |
+
"step": 240
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.03,
|
| 307 |
+
"learning_rate": 0.00012079116908177593,
|
| 308 |
+
"loss": 1.3342,
|
| 309 |
+
"step": 245
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.03,
|
| 313 |
+
"learning_rate": 0.00011736481776669306,
|
| 314 |
+
"loss": 1.2907,
|
| 315 |
+
"step": 250
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 0.03,
|
| 319 |
+
"learning_rate": 0.00011391731009600654,
|
| 320 |
+
"loss": 1.2878,
|
| 321 |
+
"step": 255
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.03,
|
| 325 |
+
"learning_rate": 0.00011045284632676536,
|
| 326 |
+
"loss": 1.2695,
|
| 327 |
+
"step": 260
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 0.03,
|
| 331 |
+
"learning_rate": 0.00010697564737441252,
|
| 332 |
+
"loss": 1.2497,
|
| 333 |
+
"step": 265
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 0.03,
|
| 337 |
+
"learning_rate": 0.00010348994967025012,
|
| 338 |
+
"loss": 1.2663,
|
| 339 |
+
"step": 270
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.03,
|
| 343 |
+
"learning_rate": 0.0001,
|
| 344 |
+
"loss": 1.2748,
|
| 345 |
+
"step": 275
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.04,
|
| 349 |
+
"learning_rate": 9.651005032974994e-05,
|
| 350 |
+
"loss": 1.2526,
|
| 351 |
+
"step": 280
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 0.04,
|
| 355 |
+
"learning_rate": 9.302435262558747e-05,
|
| 356 |
+
"loss": 1.3002,
|
| 357 |
+
"step": 285
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 0.04,
|
| 361 |
+
"learning_rate": 8.954715367323468e-05,
|
| 362 |
+
"loss": 1.2367,
|
| 363 |
+
"step": 290
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 0.04,
|
| 367 |
+
"learning_rate": 8.608268990399349e-05,
|
| 368 |
+
"loss": 1.2935,
|
| 369 |
+
"step": 295
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.04,
|
| 373 |
+
"learning_rate": 8.263518223330697e-05,
|
| 374 |
+
"loss": 1.3007,
|
| 375 |
+
"step": 300
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 0.04,
|
| 379 |
+
"learning_rate": 7.920883091822408e-05,
|
| 380 |
+
"loss": 1.3359,
|
| 381 |
+
"step": 305
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.04,
|
| 385 |
+
"learning_rate": 7.580781044003324e-05,
|
| 386 |
+
"loss": 1.2743,
|
| 387 |
+
"step": 310
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.04,
|
| 391 |
+
"learning_rate": 7.243626441830009e-05,
|
| 392 |
+
"loss": 1.2772,
|
| 393 |
+
"step": 315
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.04,
|
| 397 |
+
"learning_rate": 6.909830056250527e-05,
|
| 398 |
+
"loss": 1.2885,
|
| 399 |
+
"step": 320
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.04,
|
| 403 |
+
"learning_rate": 6.579798566743314e-05,
|
| 404 |
+
"loss": 1.2912,
|
| 405 |
+
"step": 325
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 0.04,
|
| 409 |
+
"learning_rate": 6.25393406584088e-05,
|
| 410 |
+
"loss": 1.2306,
|
| 411 |
+
"step": 330
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 0.04,
|
| 415 |
+
"learning_rate": 5.9326335692419995e-05,
|
| 416 |
+
"loss": 1.3064,
|
| 417 |
+
"step": 335
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.04,
|
| 421 |
+
"learning_rate": 5.616288532109225e-05,
|
| 422 |
+
"loss": 1.297,
|
| 423 |
+
"step": 340
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.04,
|
| 427 |
+
"learning_rate": 5.305284372141095e-05,
|
| 428 |
+
"loss": 1.2715,
|
| 429 |
+
"step": 345
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.04,
|
| 433 |
+
"learning_rate": 5.000000000000002e-05,
|
| 434 |
+
"loss": 1.2466,
|
| 435 |
+
"step": 350
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 0.04,
|
| 439 |
+
"learning_rate": 4.700807357667952e-05,
|
| 440 |
+
"loss": 1.2669,
|
| 441 |
+
"step": 355
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.05,
|
| 445 |
+
"learning_rate": 4.4080709652925336e-05,
|
| 446 |
+
"loss": 1.249,
|
| 447 |
+
"step": 360
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 0.05,
|
| 451 |
+
"learning_rate": 4.12214747707527e-05,
|
| 452 |
+
"loss": 1.235,
|
| 453 |
+
"step": 365
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 0.05,
|
| 457 |
+
"learning_rate": 3.843385246743417e-05,
|
| 458 |
+
"loss": 1.2891,
|
| 459 |
+
"step": 370
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.05,
|
| 463 |
+
"learning_rate": 3.5721239031346066e-05,
|
| 464 |
+
"loss": 1.2281,
|
| 465 |
+
"step": 375
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.05,
|
| 469 |
+
"learning_rate": 3.308693936411421e-05,
|
| 470 |
+
"loss": 1.2338,
|
| 471 |
+
"step": 380
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.05,
|
| 475 |
+
"learning_rate": 3.053416295410026e-05,
|
| 476 |
+
"loss": 1.2941,
|
| 477 |
+
"step": 385
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 0.05,
|
| 481 |
+
"learning_rate": 2.8066019966134904e-05,
|
| 482 |
+
"loss": 1.2671,
|
| 483 |
+
"step": 390
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 0.05,
|
| 487 |
+
"learning_rate": 2.5685517452260567e-05,
|
| 488 |
+
"loss": 1.3195,
|
| 489 |
+
"step": 395
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.05,
|
| 493 |
+
"learning_rate": 2.339555568810221e-05,
|
| 494 |
+
"loss": 1.2021,
|
| 495 |
+
"step": 400
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 0.05,
|
| 499 |
+
"learning_rate": 2.119892463932781e-05,
|
| 500 |
+
"loss": 1.2715,
|
| 501 |
+
"step": 405
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 0.05,
|
| 505 |
+
"learning_rate": 1.9098300562505266e-05,
|
| 506 |
+
"loss": 1.2557,
|
| 507 |
+
"step": 410
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.05,
|
| 511 |
+
"learning_rate": 1.7096242744495837e-05,
|
| 512 |
+
"loss": 1.3001,
|
| 513 |
+
"step": 415
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.05,
|
| 517 |
+
"learning_rate": 1.5195190384357404e-05,
|
| 518 |
+
"loss": 1.2931,
|
| 519 |
+
"step": 420
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.05,
|
| 523 |
+
"learning_rate": 1.339745962155613e-05,
|
| 524 |
+
"loss": 1.2374,
|
| 525 |
+
"step": 425
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 0.05,
|
| 529 |
+
"learning_rate": 1.1705240714107302e-05,
|
| 530 |
+
"loss": 1.2925,
|
| 531 |
+
"step": 430
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 0.05,
|
| 535 |
+
"learning_rate": 1.0120595370083318e-05,
|
| 536 |
+
"loss": 1.2766,
|
| 537 |
+
"step": 435
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.06,
|
| 541 |
+
"learning_rate": 8.645454235739903e-06,
|
| 542 |
+
"loss": 1.3244,
|
| 543 |
+
"step": 440
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 0.06,
|
| 547 |
+
"learning_rate": 7.281614543321269e-06,
|
| 548 |
+
"loss": 1.3404,
|
| 549 |
+
"step": 445
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.06,
|
| 553 |
+
"learning_rate": 6.030737921409169e-06,
|
| 554 |
+
"loss": 1.2886,
|
| 555 |
+
"step": 450
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.06,
|
| 559 |
+
"learning_rate": 4.8943483704846475e-06,
|
| 560 |
+
"loss": 1.3172,
|
| 561 |
+
"step": 455
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.06,
|
| 565 |
+
"learning_rate": 3.873830406168111e-06,
|
| 566 |
+
"loss": 1.2714,
|
| 567 |
+
"step": 460
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 0.06,
|
| 571 |
+
"learning_rate": 2.970427372400353e-06,
|
| 572 |
+
"loss": 1.2419,
|
| 573 |
+
"step": 465
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 0.06,
|
| 577 |
+
"learning_rate": 2.1852399266194314e-06,
|
| 578 |
+
"loss": 1.2816,
|
| 579 |
+
"step": 470
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 0.06,
|
| 583 |
+
"learning_rate": 1.5192246987791981e-06,
|
| 584 |
+
"loss": 1.252,
|
| 585 |
+
"step": 475
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.06,
|
| 589 |
+
"learning_rate": 9.731931258429638e-07,
|
| 590 |
+
"loss": 1.2246,
|
| 591 |
+
"step": 480
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.06,
|
| 595 |
+
"learning_rate": 5.478104631726711e-07,
|
| 596 |
+
"loss": 1.2728,
|
| 597 |
+
"step": 485
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.06,
|
| 601 |
+
"learning_rate": 2.4359497401758024e-07,
|
| 602 |
+
"loss": 1.2357,
|
| 603 |
+
"step": 490
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 0.06,
|
| 607 |
+
"learning_rate": 6.09172980904238e-08,
|
| 608 |
+
"loss": 1.2624,
|
| 609 |
+
"step": 495
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.06,
|
| 613 |
+
"learning_rate": 0.0,
|
| 614 |
+
"loss": 1.2922,
|
| 615 |
+
"step": 500
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 0.06,
|
| 619 |
+
"eval_loss": 1.249782919883728,
|
| 620 |
+
"eval_runtime": 1717.6564,
|
| 621 |
+
"eval_samples_per_second": 16.48,
|
| 622 |
+
"eval_steps_per_second": 1.03,
|
| 623 |
+
"step": 500
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 0.06,
|
| 627 |
+
"step": 500,
|
| 628 |
+
"total_flos": 1123799500062720.0,
|
| 629 |
+
"train_loss": 1.3271520497798919,
|
| 630 |
+
"train_runtime": 4460.9425,
|
| 631 |
+
"train_samples_per_second": 3.587,
|
| 632 |
+
"train_steps_per_second": 0.112
|
| 633 |
+
}
|
| 634 |
+
],
|
| 635 |
+
"logging_steps": 5,
|
| 636 |
+
"max_steps": 500,
|
| 637 |
+
"num_input_tokens_seen": 0,
|
| 638 |
+
"num_train_epochs": 1,
|
| 639 |
+
"save_steps": 100,
|
| 640 |
+
"total_flos": 1123799500062720.0,
|
| 641 |
+
"train_batch_size": 4,
|
| 642 |
+
"trial_name": null,
|
| 643 |
+
"trial_params": null
|
| 644 |
+
}
|