Commit
•
18011e1
1
Parent(s):
2066bc3
Model save
Browse files- README.md +62 -0
- adapter_model.safetensors +1 -1
- all_results.json +8 -0
- emissions.csv +2 -0
- runs/Sep10_19-37-45_gpu4-119-5/events.out.tfevents.1725961119.gpu4-119-5.645890.0 +2 -2
- train_results.json +8 -0
- trainer_state.json +912 -0
README.md
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- dpo
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
base_model: DUAL-GPO/phi-2-dpo-chatml-merged
|
| 9 |
+
model-index:
|
| 10 |
+
- name: phi-2-dpo-chatml-lora-10k-30k-i1
|
| 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 |
+
# phi-2-dpo-chatml-lora-10k-30k-i1
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [DUAL-GPO/phi-2-dpo-chatml-merged](https://huggingface.co/DUAL-GPO/phi-2-dpo-chatml-merged) on the None dataset.
|
| 20 |
+
|
| 21 |
+
## Model description
|
| 22 |
+
|
| 23 |
+
More information needed
|
| 24 |
+
|
| 25 |
+
## Intended uses & limitations
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Training and evaluation data
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training procedure
|
| 34 |
+
|
| 35 |
+
### Training hyperparameters
|
| 36 |
+
|
| 37 |
+
The following hyperparameters were used during training:
|
| 38 |
+
- learning_rate: 5e-06
|
| 39 |
+
- train_batch_size: 4
|
| 40 |
+
- eval_batch_size: 4
|
| 41 |
+
- seed: 42
|
| 42 |
+
- distributed_type: multi-GPU
|
| 43 |
+
- num_devices: 2
|
| 44 |
+
- gradient_accumulation_steps: 4
|
| 45 |
+
- total_train_batch_size: 32
|
| 46 |
+
- total_eval_batch_size: 8
|
| 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: 1
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
|
| 55 |
+
|
| 56 |
+
### Framework versions
|
| 57 |
+
|
| 58 |
+
- PEFT 0.7.1
|
| 59 |
+
- Transformers 4.36.2
|
| 60 |
+
- Pytorch 2.1.2
|
| 61 |
+
- Datasets 2.14.6
|
| 62 |
+
- 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 sha256:394dede7436661e7279fe5fd01776c0c97617bcda04ea9e7abff7d7021168d62
|
| 3 |
size 335579632
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.3321857452392578,
|
| 4 |
+
"train_runtime": 4002.829,
|
| 5 |
+
"train_samples": 20000,
|
| 6 |
+
"train_samples_per_second": 4.996,
|
| 7 |
+
"train_steps_per_second": 0.156
|
| 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 |
+
2024-09-10T20:45:22,codecarbon,790004d7-e26d-4cd8-9e3a-8db0ae375d65,4002.8326046466827,0.0034287817610870923,8.565888458854851e-07,42.5,652.966,188.74309015274048,0.04725510442488723,1.1859335456844342,0.20916153902987214,1.4423501891391957,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/Sep10_19-37-45_gpu4-119-5/events.out.tfevents.1725961119.gpu4-119-5.645890.0
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 sha256:0cbec3983323eb4279f09c7fe8a0308e7d20d4b1c062d8f51dd54d8ce8c029d6
|
| 3 |
+
size 25576
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.3321857452392578,
|
| 4 |
+
"train_runtime": 4002.829,
|
| 5 |
+
"train_samples": 20000,
|
| 6 |
+
"train_samples_per_second": 4.996,
|
| 7 |
+
"train_steps_per_second": 0.156
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,912 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 625,
|
| 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": 7.936507936507937e-08,
|
| 14 |
+
"logits/chosen": 0.1349913775920868,
|
| 15 |
+
"logits/rejected": 0.1167084276676178,
|
| 16 |
+
"logps/chosen": -428.1781005859375,
|
| 17 |
+
"logps/rejected": -484.54522705078125,
|
| 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": 0.02,
|
| 27 |
+
"learning_rate": 7.936507936507937e-07,
|
| 28 |
+
"logits/chosen": 0.2796071171760559,
|
| 29 |
+
"logits/rejected": 0.24151457846164703,
|
| 30 |
+
"logps/chosen": -499.6032409667969,
|
| 31 |
+
"logps/rejected": -473.9646301269531,
|
| 32 |
+
"loss": 0.6932,
|
| 33 |
+
"rewards/accuracies": 0.4027777910232544,
|
| 34 |
+
"rewards/chosen": 2.6125486328965053e-05,
|
| 35 |
+
"rewards/margins": 5.101554052089341e-05,
|
| 36 |
+
"rewards/rejected": -2.489005055394955e-05,
|
| 37 |
+
"step": 10
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.03,
|
| 41 |
+
"learning_rate": 1.5873015873015873e-06,
|
| 42 |
+
"logits/chosen": 0.2518273890018463,
|
| 43 |
+
"logits/rejected": 0.32998543977737427,
|
| 44 |
+
"logps/chosen": -489.4015197753906,
|
| 45 |
+
"logps/rejected": -479.5149841308594,
|
| 46 |
+
"loss": 0.6931,
|
| 47 |
+
"rewards/accuracies": 0.4437499940395355,
|
| 48 |
+
"rewards/chosen": -0.0016528337728232145,
|
| 49 |
+
"rewards/margins": 0.00016125899855978787,
|
| 50 |
+
"rewards/rejected": -0.001814092742279172,
|
| 51 |
+
"step": 20
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.05,
|
| 55 |
+
"learning_rate": 2.380952380952381e-06,
|
| 56 |
+
"logits/chosen": 0.18698568642139435,
|
| 57 |
+
"logits/rejected": 0.21475109457969666,
|
| 58 |
+
"logps/chosen": -501.27581787109375,
|
| 59 |
+
"logps/rejected": -522.3826293945312,
|
| 60 |
+
"loss": 0.6927,
|
| 61 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 62 |
+
"rewards/chosen": -0.008201382122933865,
|
| 63 |
+
"rewards/margins": 0.0008340775966644287,
|
| 64 |
+
"rewards/rejected": -0.009035459719598293,
|
| 65 |
+
"step": 30
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.06,
|
| 69 |
+
"learning_rate": 3.1746031746031746e-06,
|
| 70 |
+
"logits/chosen": 0.20552173256874084,
|
| 71 |
+
"logits/rejected": 0.18397466838359833,
|
| 72 |
+
"logps/chosen": -533.6796875,
|
| 73 |
+
"logps/rejected": -549.6390380859375,
|
| 74 |
+
"loss": 0.6917,
|
| 75 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 76 |
+
"rewards/chosen": -0.024371124804019928,
|
| 77 |
+
"rewards/margins": 0.0025360379368066788,
|
| 78 |
+
"rewards/rejected": -0.026907164603471756,
|
| 79 |
+
"step": 40
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.08,
|
| 83 |
+
"learning_rate": 3.968253968253968e-06,
|
| 84 |
+
"logits/chosen": 0.2108609676361084,
|
| 85 |
+
"logits/rejected": 0.2548374831676483,
|
| 86 |
+
"logps/chosen": -554.6732177734375,
|
| 87 |
+
"logps/rejected": -545.2399291992188,
|
| 88 |
+
"loss": 0.6896,
|
| 89 |
+
"rewards/accuracies": 0.5375000238418579,
|
| 90 |
+
"rewards/chosen": -0.0501377210021019,
|
| 91 |
+
"rewards/margins": 0.006772263441234827,
|
| 92 |
+
"rewards/rejected": -0.05690998584032059,
|
| 93 |
+
"step": 50
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.1,
|
| 97 |
+
"learning_rate": 4.761904761904762e-06,
|
| 98 |
+
"logits/chosen": 0.19314441084861755,
|
| 99 |
+
"logits/rejected": 0.27108246088027954,
|
| 100 |
+
"logps/chosen": -606.186767578125,
|
| 101 |
+
"logps/rejected": -608.7874145507812,
|
| 102 |
+
"loss": 0.6876,
|
| 103 |
+
"rewards/accuracies": 0.5375000238418579,
|
| 104 |
+
"rewards/chosen": -0.0944763794541359,
|
| 105 |
+
"rewards/margins": 0.008413041010499,
|
| 106 |
+
"rewards/rejected": -0.10288941860198975,
|
| 107 |
+
"step": 60
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.11,
|
| 111 |
+
"learning_rate": 4.998086282661188e-06,
|
| 112 |
+
"logits/chosen": 0.16735275089740753,
|
| 113 |
+
"logits/rejected": 0.17189382016658783,
|
| 114 |
+
"logps/chosen": -552.279296875,
|
| 115 |
+
"logps/rejected": -591.57421875,
|
| 116 |
+
"loss": 0.6852,
|
| 117 |
+
"rewards/accuracies": 0.46875,
|
| 118 |
+
"rewards/chosen": -0.11828555911779404,
|
| 119 |
+
"rewards/margins": 0.01398100983351469,
|
| 120 |
+
"rewards/rejected": -0.1322665959596634,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.13,
|
| 125 |
+
"learning_rate": 4.988720025682995e-06,
|
| 126 |
+
"logits/chosen": 0.10864056646823883,
|
| 127 |
+
"logits/rejected": 0.14530186355113983,
|
| 128 |
+
"logps/chosen": -644.5869140625,
|
| 129 |
+
"logps/rejected": -674.5743408203125,
|
| 130 |
+
"loss": 0.6828,
|
| 131 |
+
"rewards/accuracies": 0.5562499761581421,
|
| 132 |
+
"rewards/chosen": -0.13946327567100525,
|
| 133 |
+
"rewards/margins": 0.019074594601988792,
|
| 134 |
+
"rewards/rejected": -0.1585378646850586,
|
| 135 |
+
"step": 80
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.14,
|
| 139 |
+
"learning_rate": 4.9715789537359126e-06,
|
| 140 |
+
"logits/chosen": 0.07046853750944138,
|
| 141 |
+
"logits/rejected": 0.1207747831940651,
|
| 142 |
+
"logps/chosen": -684.7418212890625,
|
| 143 |
+
"logps/rejected": -712.235107421875,
|
| 144 |
+
"loss": 0.6792,
|
| 145 |
+
"rewards/accuracies": 0.612500011920929,
|
| 146 |
+
"rewards/chosen": -0.1731593906879425,
|
| 147 |
+
"rewards/margins": 0.031467147171497345,
|
| 148 |
+
"rewards/rejected": -0.20462656021118164,
|
| 149 |
+
"step": 90
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.16,
|
| 153 |
+
"learning_rate": 4.946716615897932e-06,
|
| 154 |
+
"logits/chosen": 0.0422443263232708,
|
| 155 |
+
"logits/rejected": 0.07933792471885681,
|
| 156 |
+
"logps/chosen": -741.9517211914062,
|
| 157 |
+
"logps/rejected": -768.1805419921875,
|
| 158 |
+
"loss": 0.6793,
|
| 159 |
+
"rewards/accuracies": 0.5249999761581421,
|
| 160 |
+
"rewards/chosen": -0.2586411237716675,
|
| 161 |
+
"rewards/margins": 0.030019978061318398,
|
| 162 |
+
"rewards/rejected": -0.28866106271743774,
|
| 163 |
+
"step": 100
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.18,
|
| 167 |
+
"learning_rate": 4.9142106826480114e-06,
|
| 168 |
+
"logits/chosen": 0.01217989344149828,
|
| 169 |
+
"logits/rejected": 0.036614928394556046,
|
| 170 |
+
"logps/chosen": -704.1077880859375,
|
| 171 |
+
"logps/rejected": -766.5753173828125,
|
| 172 |
+
"loss": 0.6816,
|
| 173 |
+
"rewards/accuracies": 0.59375,
|
| 174 |
+
"rewards/chosen": -0.24351103603839874,
|
| 175 |
+
"rewards/margins": 0.025916840881109238,
|
| 176 |
+
"rewards/rejected": -0.2694278359413147,
|
| 177 |
+
"step": 110
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.19,
|
| 181 |
+
"learning_rate": 4.874162703221823e-06,
|
| 182 |
+
"logits/chosen": 0.04041505232453346,
|
| 183 |
+
"logits/rejected": 0.0663461834192276,
|
| 184 |
+
"logps/chosen": -756.8778686523438,
|
| 185 |
+
"logps/rejected": -816.00244140625,
|
| 186 |
+
"loss": 0.6726,
|
| 187 |
+
"rewards/accuracies": 0.5375000238418579,
|
| 188 |
+
"rewards/chosen": -0.27453333139419556,
|
| 189 |
+
"rewards/margins": 0.04802945256233215,
|
| 190 |
+
"rewards/rejected": -0.3225627541542053,
|
| 191 |
+
"step": 120
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.21,
|
| 195 |
+
"learning_rate": 4.826697788369752e-06,
|
| 196 |
+
"logits/chosen": -0.03461530804634094,
|
| 197 |
+
"logits/rejected": 0.08036826550960541,
|
| 198 |
+
"logps/chosen": -926.38720703125,
|
| 199 |
+
"logps/rejected": -1025.3724365234375,
|
| 200 |
+
"loss": 0.6617,
|
| 201 |
+
"rewards/accuracies": 0.612500011920929,
|
| 202 |
+
"rewards/chosen": -0.422262579202652,
|
| 203 |
+
"rewards/margins": 0.0845782682299614,
|
| 204 |
+
"rewards/rejected": -0.5068408250808716,
|
| 205 |
+
"step": 130
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.22,
|
| 209 |
+
"learning_rate": 4.7719642195082224e-06,
|
| 210 |
+
"logits/chosen": 0.032650940120220184,
|
| 211 |
+
"logits/rejected": 0.02779592201113701,
|
| 212 |
+
"logps/chosen": -852.9603271484375,
|
| 213 |
+
"logps/rejected": -985.2545166015625,
|
| 214 |
+
"loss": 0.6576,
|
| 215 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 216 |
+
"rewards/chosen": -0.36906754970550537,
|
| 217 |
+
"rewards/margins": 0.09497459977865219,
|
| 218 |
+
"rewards/rejected": -0.46404212713241577,
|
| 219 |
+
"step": 140
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.24,
|
| 223 |
+
"learning_rate": 4.710132985485355e-06,
|
| 224 |
+
"logits/chosen": 0.05756285786628723,
|
| 225 |
+
"logits/rejected": 0.1693125069141388,
|
| 226 |
+
"logps/chosen": -1006.541015625,
|
| 227 |
+
"logps/rejected": -1214.77587890625,
|
| 228 |
+
"loss": 0.6513,
|
| 229 |
+
"rewards/accuracies": 0.581250011920929,
|
| 230 |
+
"rewards/chosen": -0.5491369962692261,
|
| 231 |
+
"rewards/margins": 0.171253964304924,
|
| 232 |
+
"rewards/rejected": -0.7203909158706665,
|
| 233 |
+
"step": 150
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.26,
|
| 237 |
+
"learning_rate": 4.641397248408122e-06,
|
| 238 |
+
"logits/chosen": 0.25534456968307495,
|
| 239 |
+
"logits/rejected": 0.15446144342422485,
|
| 240 |
+
"logps/chosen": -1380.378173828125,
|
| 241 |
+
"logps/rejected": -1509.271728515625,
|
| 242 |
+
"loss": 0.6491,
|
| 243 |
+
"rewards/accuracies": 0.543749988079071,
|
| 244 |
+
"rewards/chosen": -0.8604512214660645,
|
| 245 |
+
"rewards/margins": 0.11827781051397324,
|
| 246 |
+
"rewards/rejected": -0.9787289500236511,
|
| 247 |
+
"step": 160
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.27,
|
| 251 |
+
"learning_rate": 4.5659717401997655e-06,
|
| 252 |
+
"logits/chosen": 0.11406876146793365,
|
| 253 |
+
"logits/rejected": 0.2429320365190506,
|
| 254 |
+
"logps/chosen": -1162.3896484375,
|
| 255 |
+
"logps/rejected": -1300.2310791015625,
|
| 256 |
+
"loss": 0.6593,
|
| 257 |
+
"rewards/accuracies": 0.550000011920929,
|
| 258 |
+
"rewards/chosen": -0.6630619168281555,
|
| 259 |
+
"rewards/margins": 0.1277274191379547,
|
| 260 |
+
"rewards/rejected": -0.7907893061637878,
|
| 261 |
+
"step": 170
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.29,
|
| 265 |
+
"learning_rate": 4.4840920917726425e-06,
|
| 266 |
+
"logits/chosen": 0.1443721204996109,
|
| 267 |
+
"logits/rejected": 0.24939250946044922,
|
| 268 |
+
"logps/chosen": -1037.646240234375,
|
| 269 |
+
"logps/rejected": -1174.6041259765625,
|
| 270 |
+
"loss": 0.6464,
|
| 271 |
+
"rewards/accuracies": 0.5625,
|
| 272 |
+
"rewards/chosen": -0.5330273509025574,
|
| 273 |
+
"rewards/margins": 0.12969453632831573,
|
| 274 |
+
"rewards/rejected": -0.6627219915390015,
|
| 275 |
+
"step": 180
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.3,
|
| 279 |
+
"learning_rate": 4.396014096912182e-06,
|
| 280 |
+
"logits/chosen": 0.1525859534740448,
|
| 281 |
+
"logits/rejected": 0.20368242263793945,
|
| 282 |
+
"logps/chosen": -1222.385498046875,
|
| 283 |
+
"logps/rejected": -1366.5484619140625,
|
| 284 |
+
"loss": 0.6629,
|
| 285 |
+
"rewards/accuracies": 0.625,
|
| 286 |
+
"rewards/chosen": -0.7235668897628784,
|
| 287 |
+
"rewards/margins": 0.11871180683374405,
|
| 288 |
+
"rewards/rejected": -0.8422788381576538,
|
| 289 |
+
"step": 190
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.32,
|
| 293 |
+
"learning_rate": 4.302012913171584e-06,
|
| 294 |
+
"logits/chosen": 0.20741760730743408,
|
| 295 |
+
"logits/rejected": 0.23669476807117462,
|
| 296 |
+
"logps/chosen": -1129.2685546875,
|
| 297 |
+
"logps/rejected": -1311.441650390625,
|
| 298 |
+
"loss": 0.6541,
|
| 299 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 300 |
+
"rewards/chosen": -0.6570686101913452,
|
| 301 |
+
"rewards/margins": 0.16556376218795776,
|
| 302 |
+
"rewards/rejected": -0.8226323127746582,
|
| 303 |
+
"step": 200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.34,
|
| 307 |
+
"learning_rate": 4.202382202273702e-06,
|
| 308 |
+
"logits/chosen": 0.20800559222698212,
|
| 309 |
+
"logits/rejected": 0.2900869846343994,
|
| 310 |
+
"logps/chosen": -1247.2933349609375,
|
| 311 |
+
"logps/rejected": -1375.472900390625,
|
| 312 |
+
"loss": 0.6643,
|
| 313 |
+
"rewards/accuracies": 0.643750011920929,
|
| 314 |
+
"rewards/chosen": -0.6747683882713318,
|
| 315 |
+
"rewards/margins": 0.1478392481803894,
|
| 316 |
+
"rewards/rejected": -0.8226076364517212,
|
| 317 |
+
"step": 210
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.35,
|
| 321 |
+
"learning_rate": 4.097433212705492e-06,
|
| 322 |
+
"logits/chosen": 0.1988043189048767,
|
| 323 |
+
"logits/rejected": 0.2728360593318939,
|
| 324 |
+
"logps/chosen": -1089.7906494140625,
|
| 325 |
+
"logps/rejected": -1184.677490234375,
|
| 326 |
+
"loss": 0.6588,
|
| 327 |
+
"rewards/accuracies": 0.543749988079071,
|
| 328 |
+
"rewards/chosen": -0.5955246686935425,
|
| 329 |
+
"rewards/margins": 0.07477297633886337,
|
| 330 |
+
"rewards/rejected": -0.6702976226806641,
|
| 331 |
+
"step": 220
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.37,
|
| 335 |
+
"learning_rate": 3.987493807371033e-06,
|
| 336 |
+
"logits/chosen": 0.2314763069152832,
|
| 337 |
+
"logits/rejected": 0.34903794527053833,
|
| 338 |
+
"logps/chosen": -1126.2254638671875,
|
| 339 |
+
"logps/rejected": -1295.7572021484375,
|
| 340 |
+
"loss": 0.6408,
|
| 341 |
+
"rewards/accuracies": 0.625,
|
| 342 |
+
"rewards/chosen": -0.6263796091079712,
|
| 343 |
+
"rewards/margins": 0.1536393165588379,
|
| 344 |
+
"rewards/rejected": -0.7800189852714539,
|
| 345 |
+
"step": 230
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.38,
|
| 349 |
+
"learning_rate": 3.872907439340758e-06,
|
| 350 |
+
"logits/chosen": 0.24221527576446533,
|
| 351 |
+
"logits/rejected": 0.3353860080242157,
|
| 352 |
+
"logps/chosen": -1068.5975341796875,
|
| 353 |
+
"logps/rejected": -1235.1514892578125,
|
| 354 |
+
"loss": 0.6518,
|
| 355 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 356 |
+
"rewards/chosen": -0.6160915493965149,
|
| 357 |
+
"rewards/margins": 0.1515791118144989,
|
| 358 |
+
"rewards/rejected": -0.7676706910133362,
|
| 359 |
+
"step": 240
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.4,
|
| 363 |
+
"learning_rate": 3.75403207889666e-06,
|
| 364 |
+
"logits/chosen": 0.2971830368041992,
|
| 365 |
+
"logits/rejected": 0.34285473823547363,
|
| 366 |
+
"logps/chosen": -1286.203369140625,
|
| 367 |
+
"logps/rejected": -1462.2490234375,
|
| 368 |
+
"loss": 0.647,
|
| 369 |
+
"rewards/accuracies": 0.612500011920929,
|
| 370 |
+
"rewards/chosen": -0.7703440189361572,
|
| 371 |
+
"rewards/margins": 0.16007237136363983,
|
| 372 |
+
"rewards/rejected": -0.9304164052009583,
|
| 373 |
+
"step": 250
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.42,
|
| 377 |
+
"learning_rate": 3.631239095225417e-06,
|
| 378 |
+
"logits/chosen": 0.22825905680656433,
|
| 379 |
+
"logits/rejected": 0.32187604904174805,
|
| 380 |
+
"logps/chosen": -1112.8629150390625,
|
| 381 |
+
"logps/rejected": -1291.74072265625,
|
| 382 |
+
"loss": 0.64,
|
| 383 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 384 |
+
"rewards/chosen": -0.6394481658935547,
|
| 385 |
+
"rewards/margins": 0.16354107856750488,
|
| 386 |
+
"rewards/rejected": -0.8029893040657043,
|
| 387 |
+
"step": 260
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.43,
|
| 391 |
+
"learning_rate": 3.5049120962530608e-06,
|
| 392 |
+
"logits/chosen": 0.32051190733909607,
|
| 393 |
+
"logits/rejected": 0.2698783576488495,
|
| 394 |
+
"logps/chosen": -1120.5859375,
|
| 395 |
+
"logps/rejected": -1305.715087890625,
|
| 396 |
+
"loss": 0.6356,
|
| 397 |
+
"rewards/accuracies": 0.5625,
|
| 398 |
+
"rewards/chosen": -0.6476856470108032,
|
| 399 |
+
"rewards/margins": 0.15380369126796722,
|
| 400 |
+
"rewards/rejected": -0.8014893531799316,
|
| 401 |
+
"step": 270
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.45,
|
| 405 |
+
"learning_rate": 3.3754457302455464e-06,
|
| 406 |
+
"logits/chosen": 0.23477685451507568,
|
| 407 |
+
"logits/rejected": 0.43997740745544434,
|
| 408 |
+
"logps/chosen": -1300.92724609375,
|
| 409 |
+
"logps/rejected": -1495.2685546875,
|
| 410 |
+
"loss": 0.6537,
|
| 411 |
+
"rewards/accuracies": 0.5062500238418579,
|
| 412 |
+
"rewards/chosen": -0.8141768574714661,
|
| 413 |
+
"rewards/margins": 0.17344731092453003,
|
| 414 |
+
"rewards/rejected": -0.9876241683959961,
|
| 415 |
+
"step": 280
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.46,
|
| 419 |
+
"learning_rate": 3.2432444529190714e-06,
|
| 420 |
+
"logits/chosen": 0.38444215059280396,
|
| 421 |
+
"logits/rejected": 0.43280649185180664,
|
| 422 |
+
"logps/chosen": -1254.042724609375,
|
| 423 |
+
"logps/rejected": -1498.3414306640625,
|
| 424 |
+
"loss": 0.6399,
|
| 425 |
+
"rewards/accuracies": 0.550000011920929,
|
| 426 |
+
"rewards/chosen": -0.7942158579826355,
|
| 427 |
+
"rewards/margins": 0.20459000766277313,
|
| 428 |
+
"rewards/rejected": -0.9988058805465698,
|
| 429 |
+
"step": 290
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.48,
|
| 433 |
+
"learning_rate": 3.1087212639117057e-06,
|
| 434 |
+
"logits/chosen": 0.35266411304473877,
|
| 435 |
+
"logits/rejected": 0.5088330507278442,
|
| 436 |
+
"logps/chosen": -1342.4091796875,
|
| 437 |
+
"logps/rejected": -1464.3389892578125,
|
| 438 |
+
"loss": 0.6541,
|
| 439 |
+
"rewards/accuracies": 0.543749988079071,
|
| 440 |
+
"rewards/chosen": -0.8787577748298645,
|
| 441 |
+
"rewards/margins": 0.11491602659225464,
|
| 442 |
+
"rewards/rejected": -0.9936738014221191,
|
| 443 |
+
"step": 300
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.5,
|
| 447 |
+
"learning_rate": 2.9722964165636263e-06,
|
| 448 |
+
"logits/chosen": 0.3081873953342438,
|
| 449 |
+
"logits/rejected": 0.4749964773654938,
|
| 450 |
+
"logps/chosen": -1241.6292724609375,
|
| 451 |
+
"logps/rejected": -1437.963134765625,
|
| 452 |
+
"loss": 0.6523,
|
| 453 |
+
"rewards/accuracies": 0.5249999761581421,
|
| 454 |
+
"rewards/chosen": -0.7512220144271851,
|
| 455 |
+
"rewards/margins": 0.16275373101234436,
|
| 456 |
+
"rewards/rejected": -0.913975715637207,
|
| 457 |
+
"step": 310
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.51,
|
| 461 |
+
"learning_rate": 2.8343961050366275e-06,
|
| 462 |
+
"logits/chosen": 0.3727472424507141,
|
| 463 |
+
"logits/rejected": 0.41282615065574646,
|
| 464 |
+
"logps/chosen": -1198.703857421875,
|
| 465 |
+
"logps/rejected": -1467.189208984375,
|
| 466 |
+
"loss": 0.6399,
|
| 467 |
+
"rewards/accuracies": 0.612500011920929,
|
| 468 |
+
"rewards/chosen": -0.719472348690033,
|
| 469 |
+
"rewards/margins": 0.23017852008342743,
|
| 470 |
+
"rewards/rejected": -0.9496508836746216,
|
| 471 |
+
"step": 320
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.53,
|
| 475 |
+
"learning_rate": 2.695451132874385e-06,
|
| 476 |
+
"logits/chosen": 0.3753197193145752,
|
| 477 |
+
"logits/rejected": 0.4588402211666107,
|
| 478 |
+
"logps/chosen": -1306.0467529296875,
|
| 479 |
+
"logps/rejected": -1423.9765625,
|
| 480 |
+
"loss": 0.6339,
|
| 481 |
+
"rewards/accuracies": 0.5375000238418579,
|
| 482 |
+
"rewards/chosen": -0.821855902671814,
|
| 483 |
+
"rewards/margins": 0.15302011370658875,
|
| 484 |
+
"rewards/rejected": -0.9748760461807251,
|
| 485 |
+
"step": 330
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.54,
|
| 489 |
+
"learning_rate": 2.5558955671628964e-06,
|
| 490 |
+
"logits/chosen": 0.32516294717788696,
|
| 491 |
+
"logits/rejected": 0.49244314432144165,
|
| 492 |
+
"logps/chosen": -1465.662353515625,
|
| 493 |
+
"logps/rejected": -1645.747314453125,
|
| 494 |
+
"loss": 0.6415,
|
| 495 |
+
"rewards/accuracies": 0.574999988079071,
|
| 496 |
+
"rewards/chosen": -0.933737576007843,
|
| 497 |
+
"rewards/margins": 0.19395960867404938,
|
| 498 |
+
"rewards/rejected": -1.127697229385376,
|
| 499 |
+
"step": 340
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.56,
|
| 503 |
+
"learning_rate": 2.4161653824955654e-06,
|
| 504 |
+
"logits/chosen": 0.37509265542030334,
|
| 505 |
+
"logits/rejected": 0.5179189443588257,
|
| 506 |
+
"logps/chosen": -1365.8070068359375,
|
| 507 |
+
"logps/rejected": -1507.181396484375,
|
| 508 |
+
"loss": 0.639,
|
| 509 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 510 |
+
"rewards/chosen": -0.868498682975769,
|
| 511 |
+
"rewards/margins": 0.15674063563346863,
|
| 512 |
+
"rewards/rejected": -1.0252392292022705,
|
| 513 |
+
"step": 350
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.58,
|
| 517 |
+
"learning_rate": 2.2766970989791697e-06,
|
| 518 |
+
"logits/chosen": 0.3914629817008972,
|
| 519 |
+
"logits/rejected": 0.39266401529312134,
|
| 520 |
+
"logps/chosen": -1348.240966796875,
|
| 521 |
+
"logps/rejected": -1593.037353515625,
|
| 522 |
+
"loss": 0.6485,
|
| 523 |
+
"rewards/accuracies": 0.612500011920929,
|
| 524 |
+
"rewards/chosen": -0.8051446080207825,
|
| 525 |
+
"rewards/margins": 0.2146397829055786,
|
| 526 |
+
"rewards/rejected": -1.0197844505310059,
|
| 527 |
+
"step": 360
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.59,
|
| 531 |
+
"learning_rate": 2.1379264185356545e-06,
|
| 532 |
+
"logits/chosen": 0.29912859201431274,
|
| 533 |
+
"logits/rejected": 0.4203578531742096,
|
| 534 |
+
"logps/chosen": -1122.2073974609375,
|
| 535 |
+
"logps/rejected": -1355.9158935546875,
|
| 536 |
+
"loss": 0.6317,
|
| 537 |
+
"rewards/accuracies": 0.550000011920929,
|
| 538 |
+
"rewards/chosen": -0.7034318447113037,
|
| 539 |
+
"rewards/margins": 0.18991844356060028,
|
| 540 |
+
"rewards/rejected": -0.8933502435684204,
|
| 541 |
+
"step": 370
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.61,
|
| 545 |
+
"learning_rate": 2.000286863759934e-06,
|
| 546 |
+
"logits/chosen": 0.37296754121780396,
|
| 547 |
+
"logits/rejected": 0.569365382194519,
|
| 548 |
+
"logps/chosen": -1219.0467529296875,
|
| 549 |
+
"logps/rejected": -1441.828857421875,
|
| 550 |
+
"loss": 0.6332,
|
| 551 |
+
"rewards/accuracies": 0.606249988079071,
|
| 552 |
+
"rewards/chosen": -0.7400010228157043,
|
| 553 |
+
"rewards/margins": 0.20997154712677002,
|
| 554 |
+
"rewards/rejected": -0.9499725103378296,
|
| 555 |
+
"step": 380
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.62,
|
| 559 |
+
"learning_rate": 1.8642084235859764e-06,
|
| 560 |
+
"logits/chosen": 0.39869484305381775,
|
| 561 |
+
"logits/rejected": 0.49166345596313477,
|
| 562 |
+
"logps/chosen": -1271.091064453125,
|
| 563 |
+
"logps/rejected": -1447.0933837890625,
|
| 564 |
+
"loss": 0.6447,
|
| 565 |
+
"rewards/accuracies": 0.6625000238418579,
|
| 566 |
+
"rewards/chosen": -0.7757032513618469,
|
| 567 |
+
"rewards/margins": 0.18741454184055328,
|
| 568 |
+
"rewards/rejected": -0.9631179571151733,
|
| 569 |
+
"step": 390
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.64,
|
| 573 |
+
"learning_rate": 1.7301162099921013e-06,
|
| 574 |
+
"logits/chosen": 0.48826533555984497,
|
| 575 |
+
"logits/rejected": 0.5182665586471558,
|
| 576 |
+
"logps/chosen": -1342.3040771484375,
|
| 577 |
+
"logps/rejected": -1515.572021484375,
|
| 578 |
+
"loss": 0.6527,
|
| 579 |
+
"rewards/accuracies": 0.643750011920929,
|
| 580 |
+
"rewards/chosen": -0.8140773773193359,
|
| 581 |
+
"rewards/margins": 0.18610136210918427,
|
| 582 |
+
"rewards/rejected": -1.0001786947250366,
|
| 583 |
+
"step": 400
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.66,
|
| 587 |
+
"learning_rate": 1.5984291299420117e-06,
|
| 588 |
+
"logits/chosen": 0.4482479989528656,
|
| 589 |
+
"logits/rejected": 0.494678258895874,
|
| 590 |
+
"logps/chosen": -1258.864501953125,
|
| 591 |
+
"logps/rejected": -1553.4383544921875,
|
| 592 |
+
"loss": 0.6073,
|
| 593 |
+
"rewards/accuracies": 0.59375,
|
| 594 |
+
"rewards/chosen": -0.7780212163925171,
|
| 595 |
+
"rewards/margins": 0.25596219301223755,
|
| 596 |
+
"rewards/rejected": -1.0339832305908203,
|
| 597 |
+
"step": 410
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.67,
|
| 601 |
+
"learning_rate": 1.4695585767104092e-06,
|
| 602 |
+
"logits/chosen": 0.4765148162841797,
|
| 603 |
+
"logits/rejected": 0.44907188415527344,
|
| 604 |
+
"logps/chosen": -1387.763427734375,
|
| 605 |
+
"logps/rejected": -1522.737548828125,
|
| 606 |
+
"loss": 0.6573,
|
| 607 |
+
"rewards/accuracies": 0.53125,
|
| 608 |
+
"rewards/chosen": -0.9270384907722473,
|
| 609 |
+
"rewards/margins": 0.10358323156833649,
|
| 610 |
+
"rewards/rejected": -1.0306217670440674,
|
| 611 |
+
"step": 420
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.69,
|
| 615 |
+
"learning_rate": 1.3439071446815452e-06,
|
| 616 |
+
"logits/chosen": 0.4347519278526306,
|
| 617 |
+
"logits/rejected": 0.5749093294143677,
|
| 618 |
+
"logps/chosen": -1321.0262451171875,
|
| 619 |
+
"logps/rejected": -1519.8450927734375,
|
| 620 |
+
"loss": 0.6397,
|
| 621 |
+
"rewards/accuracies": 0.5625,
|
| 622 |
+
"rewards/chosen": -0.8617701530456543,
|
| 623 |
+
"rewards/margins": 0.19662390649318695,
|
| 624 |
+
"rewards/rejected": -1.0583940744400024,
|
| 625 |
+
"step": 430
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.7,
|
| 629 |
+
"learning_rate": 1.2218673716356919e-06,
|
| 630 |
+
"logits/chosen": 0.4529833197593689,
|
| 631 |
+
"logits/rejected": 0.5917688608169556,
|
| 632 |
+
"logps/chosen": -1364.6546630859375,
|
| 633 |
+
"logps/rejected": -1504.9599609375,
|
| 634 |
+
"loss": 0.6405,
|
| 635 |
+
"rewards/accuracies": 0.5625,
|
| 636 |
+
"rewards/chosen": -0.8427094221115112,
|
| 637 |
+
"rewards/margins": 0.1536230742931366,
|
| 638 |
+
"rewards/rejected": -0.9963325262069702,
|
| 639 |
+
"step": 440
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.72,
|
| 643 |
+
"learning_rate": 1.103820512452661e-06,
|
| 644 |
+
"logits/chosen": 0.3574792444705963,
|
| 645 |
+
"logits/rejected": 0.4730028510093689,
|
| 646 |
+
"logps/chosen": -1273.257568359375,
|
| 647 |
+
"logps/rejected": -1469.2452392578125,
|
| 648 |
+
"loss": 0.616,
|
| 649 |
+
"rewards/accuracies": 0.6000000238418579,
|
| 650 |
+
"rewards/chosen": -0.7781795859336853,
|
| 651 |
+
"rewards/margins": 0.17071351408958435,
|
| 652 |
+
"rewards/rejected": -0.9488931894302368,
|
| 653 |
+
"step": 450
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.74,
|
| 657 |
+
"learning_rate": 9.901353480633468e-07,
|
| 658 |
+
"logits/chosen": 0.43064388632774353,
|
| 659 |
+
"logits/rejected": 0.6706501245498657,
|
| 660 |
+
"logps/chosen": -1389.450439453125,
|
| 661 |
+
"logps/rejected": -1536.8935546875,
|
| 662 |
+
"loss": 0.6465,
|
| 663 |
+
"rewards/accuracies": 0.6312500238418579,
|
| 664 |
+
"rewards/chosen": -0.8454073071479797,
|
| 665 |
+
"rewards/margins": 0.16833043098449707,
|
| 666 |
+
"rewards/rejected": -1.0137377977371216,
|
| 667 |
+
"step": 460
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.75,
|
| 671 |
+
"learning_rate": 8.811670333701544e-07,
|
| 672 |
+
"logits/chosen": 0.46944761276245117,
|
| 673 |
+
"logits/rejected": 0.5561131238937378,
|
| 674 |
+
"logps/chosen": -1286.3804931640625,
|
| 675 |
+
"logps/rejected": -1454.6707763671875,
|
| 676 |
+
"loss": 0.6409,
|
| 677 |
+
"rewards/accuracies": 0.518750011920929,
|
| 678 |
+
"rewards/chosen": -0.8223176002502441,
|
| 679 |
+
"rewards/margins": 0.14870892465114594,
|
| 680 |
+
"rewards/rejected": -0.971026599407196,
|
| 681 |
+
"step": 470
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.77,
|
| 685 |
+
"learning_rate": 7.772559877354341e-07,
|
| 686 |
+
"logits/chosen": 0.44347596168518066,
|
| 687 |
+
"logits/rejected": 0.5607708692550659,
|
| 688 |
+
"logps/chosen": -1329.2608642578125,
|
| 689 |
+
"logps/rejected": -1553.534912109375,
|
| 690 |
+
"loss": 0.6401,
|
| 691 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 692 |
+
"rewards/chosen": -0.8247483968734741,
|
| 693 |
+
"rewards/margins": 0.23265239596366882,
|
| 694 |
+
"rewards/rejected": -1.0574008226394653,
|
| 695 |
+
"step": 480
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.78,
|
| 699 |
+
"learning_rate": 6.787268315040604e-07,
|
| 700 |
+
"logits/chosen": 0.45571979880332947,
|
| 701 |
+
"logits/rejected": 0.47239336371421814,
|
| 702 |
+
"logps/chosen": -1211.712646484375,
|
| 703 |
+
"logps/rejected": -1386.2025146484375,
|
| 704 |
+
"loss": 0.6304,
|
| 705 |
+
"rewards/accuracies": 0.612500011920929,
|
| 706 |
+
"rewards/chosen": -0.7387667298316956,
|
| 707 |
+
"rewards/margins": 0.1702551543712616,
|
| 708 |
+
"rewards/rejected": -0.9090218544006348,
|
| 709 |
+
"step": 490
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.8,
|
| 713 |
+
"learning_rate": 5.858873718824829e-07,
|
| 714 |
+
"logits/chosen": 0.3857249319553375,
|
| 715 |
+
"logits/rejected": 0.5553338527679443,
|
| 716 |
+
"logps/chosen": -1262.7962646484375,
|
| 717 |
+
"logps/rejected": -1481.9622802734375,
|
| 718 |
+
"loss": 0.6333,
|
| 719 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 720 |
+
"rewards/chosen": -0.7554186582565308,
|
| 721 |
+
"rewards/margins": 0.20985445380210876,
|
| 722 |
+
"rewards/rejected": -0.9652730822563171,
|
| 723 |
+
"step": 500
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.82,
|
| 727 |
+
"learning_rate": 4.990276413423817e-07,
|
| 728 |
+
"logits/chosen": 0.4313405454158783,
|
| 729 |
+
"logits/rejected": 0.590368390083313,
|
| 730 |
+
"logps/chosen": -1306.13427734375,
|
| 731 |
+
"logps/rejected": -1447.4407958984375,
|
| 732 |
+
"loss": 0.6303,
|
| 733 |
+
"rewards/accuracies": 0.574999988079071,
|
| 734 |
+
"rewards/chosen": -0.7738032341003418,
|
| 735 |
+
"rewards/margins": 0.16366735100746155,
|
| 736 |
+
"rewards/rejected": -0.9374706149101257,
|
| 737 |
+
"step": 510
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.83,
|
| 741 |
+
"learning_rate": 4.184189915529796e-07,
|
| 742 |
+
"logits/chosen": 0.4014933109283447,
|
| 743 |
+
"logits/rejected": 0.5485681891441345,
|
| 744 |
+
"logps/chosen": -1255.651123046875,
|
| 745 |
+
"logps/rejected": -1364.13232421875,
|
| 746 |
+
"loss": 0.6527,
|
| 747 |
+
"rewards/accuracies": 0.5062500238418579,
|
| 748 |
+
"rewards/chosen": -0.7963151931762695,
|
| 749 |
+
"rewards/margins": 0.10994601249694824,
|
| 750 |
+
"rewards/rejected": -0.9062612652778625,
|
| 751 |
+
"step": 520
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.85,
|
| 755 |
+
"learning_rate": 3.4431324567258176e-07,
|
| 756 |
+
"logits/chosen": 0.48007622361183167,
|
| 757 |
+
"logits/rejected": 0.5644385814666748,
|
| 758 |
+
"logps/chosen": -1217.1119384765625,
|
| 759 |
+
"logps/rejected": -1428.557861328125,
|
| 760 |
+
"loss": 0.6382,
|
| 761 |
+
"rewards/accuracies": 0.637499988079071,
|
| 762 |
+
"rewards/chosen": -0.7597641944885254,
|
| 763 |
+
"rewards/margins": 0.2099626362323761,
|
| 764 |
+
"rewards/rejected": -0.9697268605232239,
|
| 765 |
+
"step": 530
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.86,
|
| 769 |
+
"learning_rate": 2.769419116476052e-07,
|
| 770 |
+
"logits/chosen": 0.41416773200035095,
|
| 771 |
+
"logits/rejected": 0.5963398218154907,
|
| 772 |
+
"logps/chosen": -1281.758544921875,
|
| 773 |
+
"logps/rejected": -1478.4769287109375,
|
| 774 |
+
"loss": 0.6332,
|
| 775 |
+
"rewards/accuracies": 0.637499988079071,
|
| 776 |
+
"rewards/chosen": -0.7958284020423889,
|
| 777 |
+
"rewards/margins": 0.19340744614601135,
|
| 778 |
+
"rewards/rejected": -0.9892358779907227,
|
| 779 |
+
"step": 540
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.88,
|
| 783 |
+
"learning_rate": 2.1651545897676512e-07,
|
| 784 |
+
"logits/chosen": 0.4229779839515686,
|
| 785 |
+
"logits/rejected": 0.6195346117019653,
|
| 786 |
+
"logps/chosen": -1226.513427734375,
|
| 787 |
+
"logps/rejected": -1382.2999267578125,
|
| 788 |
+
"loss": 0.6559,
|
| 789 |
+
"rewards/accuracies": 0.5625,
|
| 790 |
+
"rewards/chosen": -0.76485675573349,
|
| 791 |
+
"rewards/margins": 0.13517045974731445,
|
| 792 |
+
"rewards/rejected": -0.9000272750854492,
|
| 793 |
+
"step": 550
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.9,
|
| 797 |
+
"learning_rate": 1.6322266119983222e-07,
|
| 798 |
+
"logits/chosen": 0.46644434332847595,
|
| 799 |
+
"logits/rejected": 0.5570663213729858,
|
| 800 |
+
"logps/chosen": -1331.478271484375,
|
| 801 |
+
"logps/rejected": -1517.951416015625,
|
| 802 |
+
"loss": 0.6477,
|
| 803 |
+
"rewards/accuracies": 0.612500011920929,
|
| 804 |
+
"rewards/chosen": -0.8168514966964722,
|
| 805 |
+
"rewards/margins": 0.17577320337295532,
|
| 806 |
+
"rewards/rejected": -0.9926246404647827,
|
| 807 |
+
"step": 560
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.91,
|
| 811 |
+
"learning_rate": 1.1723000616502167e-07,
|
| 812 |
+
"logits/chosen": 0.414206326007843,
|
| 813 |
+
"logits/rejected": 0.6192079782485962,
|
| 814 |
+
"logps/chosen": -1307.5006103515625,
|
| 815 |
+
"logps/rejected": -1502.477783203125,
|
| 816 |
+
"loss": 0.6329,
|
| 817 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 818 |
+
"rewards/chosen": -0.8042691349983215,
|
| 819 |
+
"rewards/margins": 0.1876949965953827,
|
| 820 |
+
"rewards/rejected": -0.9919641613960266,
|
| 821 |
+
"step": 570
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.93,
|
| 825 |
+
"learning_rate": 7.868117591737585e-08,
|
| 826 |
+
"logits/chosen": 0.3812820017337799,
|
| 827 |
+
"logits/rejected": 0.6579981446266174,
|
| 828 |
+
"logps/chosen": -1247.164794921875,
|
| 829 |
+
"logps/rejected": -1437.7857666015625,
|
| 830 |
+
"loss": 0.643,
|
| 831 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 832 |
+
"rewards/chosen": -0.7581318616867065,
|
| 833 |
+
"rewards/margins": 0.20403996109962463,
|
| 834 |
+
"rewards/rejected": -0.9621719121932983,
|
| 835 |
+
"step": 580
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.94,
|
| 839 |
+
"learning_rate": 4.769659783295383e-08,
|
| 840 |
+
"logits/chosen": 0.38649970293045044,
|
| 841 |
+
"logits/rejected": 0.5585275888442993,
|
| 842 |
+
"logps/chosen": -1279.88818359375,
|
| 843 |
+
"logps/rejected": -1423.707763671875,
|
| 844 |
+
"loss": 0.6325,
|
| 845 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 846 |
+
"rewards/chosen": -0.7771328687667847,
|
| 847 |
+
"rewards/margins": 0.1578211635351181,
|
| 848 |
+
"rewards/rejected": -0.9349539875984192,
|
| 849 |
+
"step": 590
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.96,
|
| 853 |
+
"learning_rate": 2.4373068401120358e-08,
|
| 854 |
+
"logits/chosen": 0.5281786918640137,
|
| 855 |
+
"logits/rejected": 0.5024128556251526,
|
| 856 |
+
"logps/chosen": -1351.9422607421875,
|
| 857 |
+
"logps/rejected": -1575.468505859375,
|
| 858 |
+
"loss": 0.6309,
|
| 859 |
+
"rewards/accuracies": 0.6000000238418579,
|
| 860 |
+
"rewards/chosen": -0.7931488752365112,
|
| 861 |
+
"rewards/margins": 0.2438853532075882,
|
| 862 |
+
"rewards/rejected": -1.037034273147583,
|
| 863 |
+
"step": 600
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.98,
|
| 867 |
+
"learning_rate": 8.78345083022425e-09,
|
| 868 |
+
"logits/chosen": 0.42136329412460327,
|
| 869 |
+
"logits/rejected": 0.5088121891021729,
|
| 870 |
+
"logps/chosen": -1306.164794921875,
|
| 871 |
+
"logps/rejected": -1567.685546875,
|
| 872 |
+
"loss": 0.6363,
|
| 873 |
+
"rewards/accuracies": 0.581250011920929,
|
| 874 |
+
"rewards/chosen": -0.8058547973632812,
|
| 875 |
+
"rewards/margins": 0.24808284640312195,
|
| 876 |
+
"rewards/rejected": -1.0539376735687256,
|
| 877 |
+
"step": 610
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.99,
|
| 881 |
+
"learning_rate": 9.764474213677654e-10,
|
| 882 |
+
"logits/chosen": 0.45875900983810425,
|
| 883 |
+
"logits/rejected": 0.5520389080047607,
|
| 884 |
+
"logps/chosen": -1280.700439453125,
|
| 885 |
+
"logps/rejected": -1439.426025390625,
|
| 886 |
+
"loss": 0.6465,
|
| 887 |
+
"rewards/accuracies": 0.543749988079071,
|
| 888 |
+
"rewards/chosen": -0.7951013445854187,
|
| 889 |
+
"rewards/margins": 0.15834487974643707,
|
| 890 |
+
"rewards/rejected": -0.9534462094306946,
|
| 891 |
+
"step": 620
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.0,
|
| 895 |
+
"step": 625,
|
| 896 |
+
"total_flos": 0.0,
|
| 897 |
+
"train_loss": 0.3321857452392578,
|
| 898 |
+
"train_runtime": 4002.829,
|
| 899 |
+
"train_samples_per_second": 4.996,
|
| 900 |
+
"train_steps_per_second": 0.156
|
| 901 |
+
}
|
| 902 |
+
],
|
| 903 |
+
"logging_steps": 10,
|
| 904 |
+
"max_steps": 625,
|
| 905 |
+
"num_input_tokens_seen": 0,
|
| 906 |
+
"num_train_epochs": 1,
|
| 907 |
+
"save_steps": 100,
|
| 908 |
+
"total_flos": 0.0,
|
| 909 |
+
"train_batch_size": 4,
|
| 910 |
+
"trial_name": null,
|
| 911 |
+
"trial_params": null
|
| 912 |
+
}
|