Commit
•
e6f2538
1
Parent(s):
c7483ea
Model save
Browse files- README.md +61 -0
- adapter_model.safetensors +1 -1
- all_results.json +8 -0
- emissions.csv +2 -0
- runs/Sep12_13-58-15_gpu4-119-5/events.out.tfevents.1726113542.gpu4-119-5.2187599.0 +2 -2
- train_results.json +8 -0
- trainer_state.json +478 -0
README.md
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
tags:
|
| 4 |
+
- trl
|
| 5 |
+
- dpo
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
base_model: DUAL-GPO/phi-2-dpo-chatml-lora-40k-60k-i1-merged
|
| 8 |
+
model-index:
|
| 9 |
+
- name: phi-2-dpo-chatml-lora-40k-60k-v2-i2
|
| 10 |
+
results: []
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 14 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 15 |
+
|
| 16 |
+
# phi-2-dpo-chatml-lora-40k-60k-v2-i2
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [DUAL-GPO/phi-2-dpo-chatml-lora-40k-60k-i1-merged](https://huggingface.co/DUAL-GPO/phi-2-dpo-chatml-lora-40k-60k-i1-merged) on the None dataset.
|
| 19 |
+
|
| 20 |
+
## Model description
|
| 21 |
+
|
| 22 |
+
More information needed
|
| 23 |
+
|
| 24 |
+
## Intended uses & limitations
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Training and evaluation data
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training procedure
|
| 33 |
+
|
| 34 |
+
### Training hyperparameters
|
| 35 |
+
|
| 36 |
+
The following hyperparameters were used during training:
|
| 37 |
+
- learning_rate: 5e-06
|
| 38 |
+
- train_batch_size: 4
|
| 39 |
+
- eval_batch_size: 4
|
| 40 |
+
- seed: 42
|
| 41 |
+
- distributed_type: multi-GPU
|
| 42 |
+
- num_devices: 4
|
| 43 |
+
- gradient_accumulation_steps: 4
|
| 44 |
+
- total_train_batch_size: 64
|
| 45 |
+
- total_eval_batch_size: 16
|
| 46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
+
- lr_scheduler_type: cosine
|
| 48 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 49 |
+
- num_epochs: 1
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
|
| 54 |
+
|
| 55 |
+
### Framework versions
|
| 56 |
+
|
| 57 |
+
- PEFT 0.7.1
|
| 58 |
+
- Transformers 4.36.2
|
| 59 |
+
- Pytorch 2.1.2
|
| 60 |
+
- Datasets 2.14.6
|
| 61 |
+
- Tokenizers 0.15.2
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 335579632
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e2e93cf34072fac22f8b4dfadb65dbe0d1dc04e584f1e110bc65709794b3774e
|
| 3 |
size 335579632
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.666122229435505,
|
| 4 |
+
"train_runtime": 4198.8516,
|
| 5 |
+
"train_samples": 20000,
|
| 6 |
+
"train_samples_per_second": 4.763,
|
| 7 |
+
"train_steps_per_second": 0.074
|
| 8 |
+
}
|
emissions.csv
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
timestamp,project_name,run_id,duration,emissions,emissions_rate,cpu_power,gpu_power,ram_power,cpu_energy,gpu_energy,ram_energy,energy_consumed,country_name,country_iso_code,region,cloud_provider,cloud_region,os,python_version,codecarbon_version,cpu_count,cpu_model,gpu_count,gpu_model,longitude,latitude,ram_total_size,tracking_mode,on_cloud,pue
|
| 2 |
+
2024-09-12T15:09:01,codecarbon,fb687a3f-aab0-4b7e-bd1e-9b0b7cb3d8e0,4198.856112241745,0.0038916446920244244,9.268344968236356e-07,42.5,1098.976,188.74309015274048,0.04956925347877874,1.3680259234841827,0.21946277793713023,1.6370579549000932,Canada,CAN,quebec,,,Linux-5.15.0-84-generic-x86_64-with-glibc2.35,3.10.14,2.2.3,32,Intel(R) Xeon(R) W-3335 CPU @ 3.40GHz,4,4 x NVIDIA GeForce RTX 4090,-71.2,46.8,503.3149070739746,machine,N,1.0
|
runs/Sep12_13-58-15_gpu4-119-5/events.out.tfevents.1726113542.gpu4-119-5.2187599.0
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 sha256:9cd3849f224fee7c84d4474d67eac12d8e4e27e0d52ec3427b11da55e55f993f
|
| 3 |
+
size 25461
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.666122229435505,
|
| 4 |
+
"train_runtime": 4198.8516,
|
| 5 |
+
"train_samples": 20000,
|
| 6 |
+
"train_samples_per_second": 4.763,
|
| 7 |
+
"train_steps_per_second": 0.074
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,478 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9984,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 312,
|
| 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": 1.5625e-07,
|
| 14 |
+
"logits/chosen": 0.37485986948013306,
|
| 15 |
+
"logits/rejected": 0.6487500071525574,
|
| 16 |
+
"logps/chosen": -1078.384765625,
|
| 17 |
+
"logps/rejected": -1101.77490234375,
|
| 18 |
+
"loss": 0.6931,
|
| 19 |
+
"rewards/accuracies": 0.0,
|
| 20 |
+
"rewards/chosen": 0.0,
|
| 21 |
+
"rewards/margins": 0.0,
|
| 22 |
+
"rewards/rejected": 0.0,
|
| 23 |
+
"step": 1
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.03,
|
| 27 |
+
"learning_rate": 1.5625e-06,
|
| 28 |
+
"logits/chosen": 0.47351381182670593,
|
| 29 |
+
"logits/rejected": 0.5273572206497192,
|
| 30 |
+
"logps/chosen": -1056.42822265625,
|
| 31 |
+
"logps/rejected": -1169.265625,
|
| 32 |
+
"loss": 0.6932,
|
| 33 |
+
"rewards/accuracies": 0.3958333432674408,
|
| 34 |
+
"rewards/chosen": -0.00091694132424891,
|
| 35 |
+
"rewards/margins": -0.00018613642896525562,
|
| 36 |
+
"rewards/rejected": -0.0007308049243874848,
|
| 37 |
+
"step": 10
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.06,
|
| 41 |
+
"learning_rate": 3.125e-06,
|
| 42 |
+
"logits/chosen": 0.42252635955810547,
|
| 43 |
+
"logits/rejected": 0.49473732709884644,
|
| 44 |
+
"logps/chosen": -1147.17236328125,
|
| 45 |
+
"logps/rejected": -1265.768798828125,
|
| 46 |
+
"loss": 0.6914,
|
| 47 |
+
"rewards/accuracies": 0.48124998807907104,
|
| 48 |
+
"rewards/chosen": -0.022750383242964745,
|
| 49 |
+
"rewards/margins": 0.0030057504773139954,
|
| 50 |
+
"rewards/rejected": -0.02575613185763359,
|
| 51 |
+
"step": 20
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.1,
|
| 55 |
+
"learning_rate": 4.6875000000000004e-06,
|
| 56 |
+
"logits/chosen": 0.5097017884254456,
|
| 57 |
+
"logits/rejected": 0.5685985088348389,
|
| 58 |
+
"logps/chosen": -1142.3890380859375,
|
| 59 |
+
"logps/rejected": -1274.76171875,
|
| 60 |
+
"loss": 0.6878,
|
| 61 |
+
"rewards/accuracies": 0.5,
|
| 62 |
+
"rewards/chosen": -0.09474565088748932,
|
| 63 |
+
"rewards/margins": 0.013504189439117908,
|
| 64 |
+
"rewards/rejected": -0.10824984312057495,
|
| 65 |
+
"step": 30
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.13,
|
| 69 |
+
"learning_rate": 4.989935734988098e-06,
|
| 70 |
+
"logits/chosen": 0.6262896656990051,
|
| 71 |
+
"logits/rejected": 0.5369861125946045,
|
| 72 |
+
"logps/chosen": -1379.107177734375,
|
| 73 |
+
"logps/rejected": -1572.9727783203125,
|
| 74 |
+
"loss": 0.679,
|
| 75 |
+
"rewards/accuracies": 0.5562499761581421,
|
| 76 |
+
"rewards/chosen": -0.25016888976097107,
|
| 77 |
+
"rewards/margins": 0.05141867324709892,
|
| 78 |
+
"rewards/rejected": -0.3015875816345215,
|
| 79 |
+
"step": 40
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.16,
|
| 83 |
+
"learning_rate": 4.949188496058089e-06,
|
| 84 |
+
"logits/chosen": 0.595242977142334,
|
| 85 |
+
"logits/rejected": 0.6554633378982544,
|
| 86 |
+
"logps/chosen": -1302.7606201171875,
|
| 87 |
+
"logps/rejected": -1508.60107421875,
|
| 88 |
+
"loss": 0.6637,
|
| 89 |
+
"rewards/accuracies": 0.48750001192092896,
|
| 90 |
+
"rewards/chosen": -0.35188037157058716,
|
| 91 |
+
"rewards/margins": 0.07706869393587112,
|
| 92 |
+
"rewards/rejected": -0.4289490282535553,
|
| 93 |
+
"step": 50
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.19,
|
| 97 |
+
"learning_rate": 4.8776412907378845e-06,
|
| 98 |
+
"logits/chosen": 0.6328147053718567,
|
| 99 |
+
"logits/rejected": 0.695043683052063,
|
| 100 |
+
"logps/chosen": -1562.052978515625,
|
| 101 |
+
"logps/rejected": -1646.878662109375,
|
| 102 |
+
"loss": 0.6789,
|
| 103 |
+
"rewards/accuracies": 0.4749999940395355,
|
| 104 |
+
"rewards/chosen": -0.4670296609401703,
|
| 105 |
+
"rewards/margins": 0.025544878095388412,
|
| 106 |
+
"rewards/rejected": -0.492574542760849,
|
| 107 |
+
"step": 60
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.22,
|
| 111 |
+
"learning_rate": 4.7761938666470405e-06,
|
| 112 |
+
"logits/chosen": 0.5762341618537903,
|
| 113 |
+
"logits/rejected": 0.6952670812606812,
|
| 114 |
+
"logps/chosen": -1321.309814453125,
|
| 115 |
+
"logps/rejected": -1596.348876953125,
|
| 116 |
+
"loss": 0.6667,
|
| 117 |
+
"rewards/accuracies": 0.606249988079071,
|
| 118 |
+
"rewards/chosen": -0.27966535091400146,
|
| 119 |
+
"rewards/margins": 0.1273583322763443,
|
| 120 |
+
"rewards/rejected": -0.40702366828918457,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.26,
|
| 125 |
+
"learning_rate": 4.646121984004666e-06,
|
| 126 |
+
"logits/chosen": 0.5692261457443237,
|
| 127 |
+
"logits/rejected": 0.8441296815872192,
|
| 128 |
+
"logps/chosen": -1434.124755859375,
|
| 129 |
+
"logps/rejected": -1714.9644775390625,
|
| 130 |
+
"loss": 0.6653,
|
| 131 |
+
"rewards/accuracies": 0.512499988079071,
|
| 132 |
+
"rewards/chosen": -0.38832995295524597,
|
| 133 |
+
"rewards/margins": 0.12836144864559174,
|
| 134 |
+
"rewards/rejected": -0.5166913866996765,
|
| 135 |
+
"step": 80
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.29,
|
| 139 |
+
"learning_rate": 4.4890613722044526e-06,
|
| 140 |
+
"logits/chosen": 0.5933000445365906,
|
| 141 |
+
"logits/rejected": 0.7363389730453491,
|
| 142 |
+
"logps/chosen": -1363.330810546875,
|
| 143 |
+
"logps/rejected": -1636.5830078125,
|
| 144 |
+
"loss": 0.6614,
|
| 145 |
+
"rewards/accuracies": 0.581250011920929,
|
| 146 |
+
"rewards/chosen": -0.36261868476867676,
|
| 147 |
+
"rewards/margins": 0.11317511647939682,
|
| 148 |
+
"rewards/rejected": -0.47579383850097656,
|
| 149 |
+
"step": 90
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.32,
|
| 153 |
+
"learning_rate": 4.3069871595684795e-06,
|
| 154 |
+
"logits/chosen": 0.6000683903694153,
|
| 155 |
+
"logits/rejected": 0.7250877618789673,
|
| 156 |
+
"logps/chosen": -1501.1116943359375,
|
| 157 |
+
"logps/rejected": -1688.3638916015625,
|
| 158 |
+
"loss": 0.6729,
|
| 159 |
+
"rewards/accuracies": 0.512499988079071,
|
| 160 |
+
"rewards/chosen": -0.42007994651794434,
|
| 161 |
+
"rewards/margins": 0.07729745656251907,
|
| 162 |
+
"rewards/rejected": -0.49737733602523804,
|
| 163 |
+
"step": 100
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.35,
|
| 167 |
+
"learning_rate": 4.102189034962561e-06,
|
| 168 |
+
"logits/chosen": 0.6664993166923523,
|
| 169 |
+
"logits/rejected": 0.8522599935531616,
|
| 170 |
+
"logps/chosen": -1442.220703125,
|
| 171 |
+
"logps/rejected": -1732.263671875,
|
| 172 |
+
"loss": 0.6579,
|
| 173 |
+
"rewards/accuracies": 0.581250011920929,
|
| 174 |
+
"rewards/chosen": -0.3821481168270111,
|
| 175 |
+
"rewards/margins": 0.13758106529712677,
|
| 176 |
+
"rewards/rejected": -0.5197292566299438,
|
| 177 |
+
"step": 110
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.38,
|
| 181 |
+
"learning_rate": 3.8772424536302565e-06,
|
| 182 |
+
"logits/chosen": 0.7149074077606201,
|
| 183 |
+
"logits/rejected": 0.7848154902458191,
|
| 184 |
+
"logps/chosen": -1421.830322265625,
|
| 185 |
+
"logps/rejected": -1694.971923828125,
|
| 186 |
+
"loss": 0.6525,
|
| 187 |
+
"rewards/accuracies": 0.543749988079071,
|
| 188 |
+
"rewards/chosen": -0.38273271918296814,
|
| 189 |
+
"rewards/margins": 0.12498722970485687,
|
| 190 |
+
"rewards/rejected": -0.5077199935913086,
|
| 191 |
+
"step": 120
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.42,
|
| 195 |
+
"learning_rate": 3.634976249348867e-06,
|
| 196 |
+
"logits/chosen": 0.709136962890625,
|
| 197 |
+
"logits/rejected": 0.7118825912475586,
|
| 198 |
+
"logps/chosen": -1674.842041015625,
|
| 199 |
+
"logps/rejected": -1917.5904541015625,
|
| 200 |
+
"loss": 0.6612,
|
| 201 |
+
"rewards/accuracies": 0.550000011920929,
|
| 202 |
+
"rewards/chosen": -0.5186460614204407,
|
| 203 |
+
"rewards/margins": 0.10890078544616699,
|
| 204 |
+
"rewards/rejected": -0.6275469064712524,
|
| 205 |
+
"step": 130
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.45,
|
| 209 |
+
"learning_rate": 3.3784370602033572e-06,
|
| 210 |
+
"logits/chosen": 0.6097074747085571,
|
| 211 |
+
"logits/rejected": 0.8623794317245483,
|
| 212 |
+
"logps/chosen": -1559.768798828125,
|
| 213 |
+
"logps/rejected": -1758.3902587890625,
|
| 214 |
+
"loss": 0.6673,
|
| 215 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 216 |
+
"rewards/chosen": -0.446970134973526,
|
| 217 |
+
"rewards/margins": 0.08759806305170059,
|
| 218 |
+
"rewards/rejected": -0.534568190574646,
|
| 219 |
+
"step": 140
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.48,
|
| 223 |
+
"learning_rate": 3.1108510153447352e-06,
|
| 224 |
+
"logits/chosen": 0.567806601524353,
|
| 225 |
+
"logits/rejected": 0.8320453763008118,
|
| 226 |
+
"logps/chosen": -1542.0999755859375,
|
| 227 |
+
"logps/rejected": -1677.0787353515625,
|
| 228 |
+
"loss": 0.6708,
|
| 229 |
+
"rewards/accuracies": 0.46875,
|
| 230 |
+
"rewards/chosen": -0.41464248299598694,
|
| 231 |
+
"rewards/margins": 0.05952323600649834,
|
| 232 |
+
"rewards/rejected": -0.47416573762893677,
|
| 233 |
+
"step": 150
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.51,
|
| 237 |
+
"learning_rate": 2.835583164544139e-06,
|
| 238 |
+
"logits/chosen": 0.7906177639961243,
|
| 239 |
+
"logits/rejected": 0.7841562628746033,
|
| 240 |
+
"logps/chosen": -1429.5552978515625,
|
| 241 |
+
"logps/rejected": -1619.2171630859375,
|
| 242 |
+
"loss": 0.6578,
|
| 243 |
+
"rewards/accuracies": 0.543749988079071,
|
| 244 |
+
"rewards/chosen": -0.3786751627922058,
|
| 245 |
+
"rewards/margins": 0.09149602800607681,
|
| 246 |
+
"rewards/rejected": -0.47017115354537964,
|
| 247 |
+
"step": 160
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.54,
|
| 251 |
+
"learning_rate": 2.556095160739513e-06,
|
| 252 |
+
"logits/chosen": 0.7081605195999146,
|
| 253 |
+
"logits/rejected": 0.6871576905250549,
|
| 254 |
+
"logps/chosen": -1445.623779296875,
|
| 255 |
+
"logps/rejected": -1650.127197265625,
|
| 256 |
+
"loss": 0.6577,
|
| 257 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 258 |
+
"rewards/chosen": -0.41454702615737915,
|
| 259 |
+
"rewards/margins": 0.10776009410619736,
|
| 260 |
+
"rewards/rejected": -0.5223071575164795,
|
| 261 |
+
"step": 170
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.58,
|
| 265 |
+
"learning_rate": 2.2759017277414165e-06,
|
| 266 |
+
"logits/chosen": 0.7904581427574158,
|
| 267 |
+
"logits/rejected": 0.7038453817367554,
|
| 268 |
+
"logps/chosen": -1671.363525390625,
|
| 269 |
+
"logps/rejected": -1868.4300537109375,
|
| 270 |
+
"loss": 0.6579,
|
| 271 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 272 |
+
"rewards/chosen": -0.5421901941299438,
|
| 273 |
+
"rewards/margins": 0.10026909410953522,
|
| 274 |
+
"rewards/rejected": -0.6424592733383179,
|
| 275 |
+
"step": 180
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.61,
|
| 279 |
+
"learning_rate": 1.9985264605418185e-06,
|
| 280 |
+
"logits/chosen": 0.6540366411209106,
|
| 281 |
+
"logits/rejected": 0.7285584807395935,
|
| 282 |
+
"logps/chosen": -1518.58935546875,
|
| 283 |
+
"logps/rejected": -1726.029052734375,
|
| 284 |
+
"loss": 0.652,
|
| 285 |
+
"rewards/accuracies": 0.512499988079071,
|
| 286 |
+
"rewards/chosen": -0.4512515962123871,
|
| 287 |
+
"rewards/margins": 0.09267839789390564,
|
| 288 |
+
"rewards/rejected": -0.5439299941062927,
|
| 289 |
+
"step": 190
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.64,
|
| 293 |
+
"learning_rate": 1.7274575140626318e-06,
|
| 294 |
+
"logits/chosen": 0.879226565361023,
|
| 295 |
+
"logits/rejected": 0.8555147051811218,
|
| 296 |
+
"logps/chosen": -1627.2894287109375,
|
| 297 |
+
"logps/rejected": -1928.3841552734375,
|
| 298 |
+
"loss": 0.6602,
|
| 299 |
+
"rewards/accuracies": 0.5375000238418579,
|
| 300 |
+
"rewards/chosen": -0.5301727056503296,
|
| 301 |
+
"rewards/margins": 0.161948561668396,
|
| 302 |
+
"rewards/rejected": -0.6921212673187256,
|
| 303 |
+
"step": 200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.67,
|
| 307 |
+
"learning_rate": 1.466103737583699e-06,
|
| 308 |
+
"logits/chosen": 0.6861797571182251,
|
| 309 |
+
"logits/rejected": 0.9016023874282837,
|
| 310 |
+
"logps/chosen": -1539.524658203125,
|
| 311 |
+
"logps/rejected": -1874.627685546875,
|
| 312 |
+
"loss": 0.659,
|
| 313 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 314 |
+
"rewards/chosen": -0.4866175651550293,
|
| 315 |
+
"rewards/margins": 0.17361479997634888,
|
| 316 |
+
"rewards/rejected": -0.6602323651313782,
|
| 317 |
+
"step": 210
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.7,
|
| 321 |
+
"learning_rate": 1.217751806485235e-06,
|
| 322 |
+
"logits/chosen": 0.7002454996109009,
|
| 323 |
+
"logits/rejected": 0.8032233119010925,
|
| 324 |
+
"logps/chosen": -1692.8636474609375,
|
| 325 |
+
"logps/rejected": -1974.6669921875,
|
| 326 |
+
"loss": 0.6663,
|
| 327 |
+
"rewards/accuracies": 0.5562499761581421,
|
| 328 |
+
"rewards/chosen": -0.5267156958580017,
|
| 329 |
+
"rewards/margins": 0.14205826818943024,
|
| 330 |
+
"rewards/rejected": -0.6687740087509155,
|
| 331 |
+
"step": 220
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.74,
|
| 335 |
+
"learning_rate": 9.855248903979505e-07,
|
| 336 |
+
"logits/chosen": 0.7965744733810425,
|
| 337 |
+
"logits/rejected": 0.7885487079620361,
|
| 338 |
+
"logps/chosen": -1604.53466796875,
|
| 339 |
+
"logps/rejected": -1794.8245849609375,
|
| 340 |
+
"loss": 0.6672,
|
| 341 |
+
"rewards/accuracies": 0.5625,
|
| 342 |
+
"rewards/chosen": -0.48480549454689026,
|
| 343 |
+
"rewards/margins": 0.0899248868227005,
|
| 344 |
+
"rewards/rejected": -0.574730396270752,
|
| 345 |
+
"step": 230
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.77,
|
| 349 |
+
"learning_rate": 7.723433775328385e-07,
|
| 350 |
+
"logits/chosen": 0.7397829294204712,
|
| 351 |
+
"logits/rejected": 0.8248909711837769,
|
| 352 |
+
"logps/chosen": -1530.4615478515625,
|
| 353 |
+
"logps/rejected": -1826.6988525390625,
|
| 354 |
+
"loss": 0.654,
|
| 355 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 356 |
+
"rewards/chosen": -0.41139334440231323,
|
| 357 |
+
"rewards/margins": 0.1389993578195572,
|
| 358 |
+
"rewards/rejected": -0.550392746925354,
|
| 359 |
+
"step": 240
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.8,
|
| 363 |
+
"learning_rate": 5.808881491049723e-07,
|
| 364 |
+
"logits/chosen": 0.6129003763198853,
|
| 365 |
+
"logits/rejected": 0.8366864919662476,
|
| 366 |
+
"logps/chosen": -1451.37890625,
|
| 367 |
+
"logps/rejected": -1729.635986328125,
|
| 368 |
+
"loss": 0.6728,
|
| 369 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 370 |
+
"rewards/chosen": -0.40951260924339294,
|
| 371 |
+
"rewards/margins": 0.1274668127298355,
|
| 372 |
+
"rewards/rejected": -0.5369793772697449,
|
| 373 |
+
"step": 250
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.83,
|
| 377 |
+
"learning_rate": 4.1356686569674344e-07,
|
| 378 |
+
"logits/chosen": 0.7178203463554382,
|
| 379 |
+
"logits/rejected": 0.8371836543083191,
|
| 380 |
+
"logps/chosen": -1524.3497314453125,
|
| 381 |
+
"logps/rejected": -1778.346923828125,
|
| 382 |
+
"loss": 0.6495,
|
| 383 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 384 |
+
"rewards/chosen": -0.4218766689300537,
|
| 385 |
+
"rewards/margins": 0.12465916574001312,
|
| 386 |
+
"rewards/rejected": -0.546535849571228,
|
| 387 |
+
"step": 260
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.86,
|
| 391 |
+
"learning_rate": 2.7248368952908055e-07,
|
| 392 |
+
"logits/chosen": 0.6586390733718872,
|
| 393 |
+
"logits/rejected": 0.8500372767448425,
|
| 394 |
+
"logps/chosen": -1523.057373046875,
|
| 395 |
+
"logps/rejected": -1753.3544921875,
|
| 396 |
+
"loss": 0.6641,
|
| 397 |
+
"rewards/accuracies": 0.574999988079071,
|
| 398 |
+
"rewards/chosen": -0.43091854453086853,
|
| 399 |
+
"rewards/margins": 0.11140650510787964,
|
| 400 |
+
"rewards/rejected": -0.5423250198364258,
|
| 401 |
+
"step": 270
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.9,
|
| 405 |
+
"learning_rate": 1.59412823400657e-07,
|
| 406 |
+
"logits/chosen": 0.7076243162155151,
|
| 407 |
+
"logits/rejected": 0.7846710681915283,
|
| 408 |
+
"logps/chosen": -1433.0615234375,
|
| 409 |
+
"logps/rejected": -1673.491455078125,
|
| 410 |
+
"loss": 0.6713,
|
| 411 |
+
"rewards/accuracies": 0.4937500059604645,
|
| 412 |
+
"rewards/chosen": -0.39409512281417847,
|
| 413 |
+
"rewards/margins": 0.11671394109725952,
|
| 414 |
+
"rewards/rejected": -0.510809063911438,
|
| 415 |
+
"step": 280
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.93,
|
| 419 |
+
"learning_rate": 7.577619905828281e-08,
|
| 420 |
+
"logits/chosen": 0.6640155911445618,
|
| 421 |
+
"logits/rejected": 0.7537108659744263,
|
| 422 |
+
"logps/chosen": -1413.724853515625,
|
| 423 |
+
"logps/rejected": -1650.0406494140625,
|
| 424 |
+
"loss": 0.6651,
|
| 425 |
+
"rewards/accuracies": 0.543749988079071,
|
| 426 |
+
"rewards/chosen": -0.38546374440193176,
|
| 427 |
+
"rewards/margins": 0.11493394523859024,
|
| 428 |
+
"rewards/rejected": -0.5003976821899414,
|
| 429 |
+
"step": 290
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.96,
|
| 433 |
+
"learning_rate": 2.262559558016325e-08,
|
| 434 |
+
"logits/chosen": 0.635712742805481,
|
| 435 |
+
"logits/rejected": 0.8228713274002075,
|
| 436 |
+
"logps/chosen": -1463.5419921875,
|
| 437 |
+
"logps/rejected": -1637.050048828125,
|
| 438 |
+
"loss": 0.6578,
|
| 439 |
+
"rewards/accuracies": 0.59375,
|
| 440 |
+
"rewards/chosen": -0.41111892461776733,
|
| 441 |
+
"rewards/margins": 0.08340780436992645,
|
| 442 |
+
"rewards/rejected": -0.4945267140865326,
|
| 443 |
+
"step": 300
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.99,
|
| 447 |
+
"learning_rate": 6.294126437336734e-10,
|
| 448 |
+
"logits/chosen": 0.6655277013778687,
|
| 449 |
+
"logits/rejected": 0.759280800819397,
|
| 450 |
+
"logps/chosen": -1619.28857421875,
|
| 451 |
+
"logps/rejected": -1727.4703369140625,
|
| 452 |
+
"loss": 0.6663,
|
| 453 |
+
"rewards/accuracies": 0.512499988079071,
|
| 454 |
+
"rewards/chosen": -0.4628829061985016,
|
| 455 |
+
"rewards/margins": 0.05492577701807022,
|
| 456 |
+
"rewards/rejected": -0.517808735370636,
|
| 457 |
+
"step": 310
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.0,
|
| 461 |
+
"step": 312,
|
| 462 |
+
"total_flos": 0.0,
|
| 463 |
+
"train_loss": 0.666122229435505,
|
| 464 |
+
"train_runtime": 4198.8516,
|
| 465 |
+
"train_samples_per_second": 4.763,
|
| 466 |
+
"train_steps_per_second": 0.074
|
| 467 |
+
}
|
| 468 |
+
],
|
| 469 |
+
"logging_steps": 10,
|
| 470 |
+
"max_steps": 312,
|
| 471 |
+
"num_input_tokens_seen": 0,
|
| 472 |
+
"num_train_epochs": 1,
|
| 473 |
+
"save_steps": 100,
|
| 474 |
+
"total_flos": 0.0,
|
| 475 |
+
"train_batch_size": 4,
|
| 476 |
+
"trial_name": null,
|
| 477 |
+
"trial_params": null
|
| 478 |
+
}
|