Model save
Browse files- README.md +62 -0
- adapter_model.safetensors +1 -1
- all_results.json +8 -0
- emissions.csv +2 -0
- runs/Sep11_17-50-45_gpu4-119-5/events.out.tfevents.1726041087.gpu4-119-5.1456867.0 +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-40k-60k-i2
|
| 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-40k-60k-i2
|
| 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:b9417072d043c9885b43545eb3534d34c49e01253c8c7a0285825cdf91f07862
|
| 3 |
size 335579632
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.6552388866424561,
|
| 4 |
+
"train_runtime": 7636.4902,
|
| 5 |
+
"train_samples": 20000,
|
| 6 |
+
"train_samples_per_second": 2.619,
|
| 7 |
+
"train_steps_per_second": 0.082
|
| 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 |
+
2024-09-11T19:58:43,codecarbon,c24f91b2-4c50-4cc2-9f13-9ed27f7eedf2,7636.492887973785,0.0050408220127666084,6.600964718640766e-07,42.5,656.961,188.74309015274048,0.0901520291388036,1.630853392398971,0.3994650808559932,2.120470502393769,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/Sep11_17-50-45_gpu4-119-5/events.out.tfevents.1726041087.gpu4-119-5.1456867.0
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 sha256:095356828adf6b25e68d31571144c591a0020f7d1ac51d9b0d7ec69003e76d4c
|
| 3 |
+
size 45087
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.6552388866424561,
|
| 4 |
+
"train_runtime": 7636.4902,
|
| 5 |
+
"train_samples": 20000,
|
| 6 |
+
"train_samples_per_second": 2.619,
|
| 7 |
+
"train_steps_per_second": 0.082
|
| 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": 0.24662643671035767,
|
| 15 |
+
"logits/rejected": 0.3199857771396637,
|
| 16 |
+
"logps/chosen": -398.9810791015625,
|
| 17 |
+
"logps/rejected": -445.0414123535156,
|
| 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": 0.2540830075740814,
|
| 29 |
+
"logits/rejected": 0.28116142749786377,
|
| 30 |
+
"logps/chosen": -492.2332763671875,
|
| 31 |
+
"logps/rejected": -497.7977294921875,
|
| 32 |
+
"loss": 0.6931,
|
| 33 |
+
"rewards/accuracies": 0.3958333432674408,
|
| 34 |
+
"rewards/chosen": -0.0001207808600156568,
|
| 35 |
+
"rewards/margins": -3.325760189909488e-05,
|
| 36 |
+
"rewards/rejected": -8.752326539251953e-05,
|
| 37 |
+
"step": 10
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.03,
|
| 41 |
+
"learning_rate": 1.5873015873015873e-06,
|
| 42 |
+
"logits/chosen": 0.19658036530017853,
|
| 43 |
+
"logits/rejected": 0.304654985666275,
|
| 44 |
+
"logps/chosen": -462.8489685058594,
|
| 45 |
+
"logps/rejected": -485.5413513183594,
|
| 46 |
+
"loss": 0.693,
|
| 47 |
+
"rewards/accuracies": 0.4375,
|
| 48 |
+
"rewards/chosen": -0.001320437528192997,
|
| 49 |
+
"rewards/margins": 0.0002517173998057842,
|
| 50 |
+
"rewards/rejected": -0.001572155044414103,
|
| 51 |
+
"step": 20
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.05,
|
| 55 |
+
"learning_rate": 2.380952380952381e-06,
|
| 56 |
+
"logits/chosen": 0.233014315366745,
|
| 57 |
+
"logits/rejected": 0.2545226514339447,
|
| 58 |
+
"logps/chosen": -516.826171875,
|
| 59 |
+
"logps/rejected": -515.098876953125,
|
| 60 |
+
"loss": 0.6921,
|
| 61 |
+
"rewards/accuracies": 0.59375,
|
| 62 |
+
"rewards/chosen": -0.008494555950164795,
|
| 63 |
+
"rewards/margins": 0.0018449639901518822,
|
| 64 |
+
"rewards/rejected": -0.010339519940316677,
|
| 65 |
+
"step": 30
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.06,
|
| 69 |
+
"learning_rate": 3.1746031746031746e-06,
|
| 70 |
+
"logits/chosen": 0.17288276553153992,
|
| 71 |
+
"logits/rejected": 0.2692878246307373,
|
| 72 |
+
"logps/chosen": -518.7865600585938,
|
| 73 |
+
"logps/rejected": -552.9195556640625,
|
| 74 |
+
"loss": 0.6906,
|
| 75 |
+
"rewards/accuracies": 0.5562499761581421,
|
| 76 |
+
"rewards/chosen": -0.03214402496814728,
|
| 77 |
+
"rewards/margins": 0.004966778215020895,
|
| 78 |
+
"rewards/rejected": -0.03711080551147461,
|
| 79 |
+
"step": 40
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.08,
|
| 83 |
+
"learning_rate": 3.968253968253968e-06,
|
| 84 |
+
"logits/chosen": 0.2141023576259613,
|
| 85 |
+
"logits/rejected": 0.2770322263240814,
|
| 86 |
+
"logps/chosen": -574.9302978515625,
|
| 87 |
+
"logps/rejected": -587.6199951171875,
|
| 88 |
+
"loss": 0.6899,
|
| 89 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 90 |
+
"rewards/chosen": -0.07075769454240799,
|
| 91 |
+
"rewards/margins": 0.009018848650157452,
|
| 92 |
+
"rewards/rejected": -0.07977654784917831,
|
| 93 |
+
"step": 50
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.1,
|
| 97 |
+
"learning_rate": 4.761904761904762e-06,
|
| 98 |
+
"logits/chosen": 0.20746545493602753,
|
| 99 |
+
"logits/rejected": 0.22541038691997528,
|
| 100 |
+
"logps/chosen": -621.58935546875,
|
| 101 |
+
"logps/rejected": -625.3094482421875,
|
| 102 |
+
"loss": 0.6895,
|
| 103 |
+
"rewards/accuracies": 0.48124998807907104,
|
| 104 |
+
"rewards/chosen": -0.10110817104578018,
|
| 105 |
+
"rewards/margins": 0.006856895051896572,
|
| 106 |
+
"rewards/rejected": -0.10796506702899933,
|
| 107 |
+
"step": 60
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.11,
|
| 111 |
+
"learning_rate": 4.998086282661188e-06,
|
| 112 |
+
"logits/chosen": 0.2066463679075241,
|
| 113 |
+
"logits/rejected": 0.218001127243042,
|
| 114 |
+
"logps/chosen": -615.7960815429688,
|
| 115 |
+
"logps/rejected": -646.2564697265625,
|
| 116 |
+
"loss": 0.6859,
|
| 117 |
+
"rewards/accuracies": 0.5,
|
| 118 |
+
"rewards/chosen": -0.1264130175113678,
|
| 119 |
+
"rewards/margins": 0.008419327437877655,
|
| 120 |
+
"rewards/rejected": -0.13483233749866486,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.13,
|
| 125 |
+
"learning_rate": 4.988720025682995e-06,
|
| 126 |
+
"logits/chosen": 0.17233335971832275,
|
| 127 |
+
"logits/rejected": 0.15302792191505432,
|
| 128 |
+
"logps/chosen": -658.7843017578125,
|
| 129 |
+
"logps/rejected": -671.2628173828125,
|
| 130 |
+
"loss": 0.6848,
|
| 131 |
+
"rewards/accuracies": 0.581250011920929,
|
| 132 |
+
"rewards/chosen": -0.12070327997207642,
|
| 133 |
+
"rewards/margins": 0.018190523609519005,
|
| 134 |
+
"rewards/rejected": -0.13889379799365997,
|
| 135 |
+
"step": 80
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.14,
|
| 139 |
+
"learning_rate": 4.9715789537359126e-06,
|
| 140 |
+
"logits/chosen": 0.10335955768823624,
|
| 141 |
+
"logits/rejected": 0.1700659692287445,
|
| 142 |
+
"logps/chosen": -580.2098999023438,
|
| 143 |
+
"logps/rejected": -629.610595703125,
|
| 144 |
+
"loss": 0.6797,
|
| 145 |
+
"rewards/accuracies": 0.5625,
|
| 146 |
+
"rewards/chosen": -0.14437471330165863,
|
| 147 |
+
"rewards/margins": 0.03023264743387699,
|
| 148 |
+
"rewards/rejected": -0.17460736632347107,
|
| 149 |
+
"step": 90
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.16,
|
| 153 |
+
"learning_rate": 4.946716615897932e-06,
|
| 154 |
+
"logits/chosen": 0.08332869410514832,
|
| 155 |
+
"logits/rejected": 0.09228050708770752,
|
| 156 |
+
"logps/chosen": -648.9441528320312,
|
| 157 |
+
"logps/rejected": -705.8828735351562,
|
| 158 |
+
"loss": 0.6796,
|
| 159 |
+
"rewards/accuracies": 0.543749988079071,
|
| 160 |
+
"rewards/chosen": -0.19636474549770355,
|
| 161 |
+
"rewards/margins": 0.02650091052055359,
|
| 162 |
+
"rewards/rejected": -0.22286565601825714,
|
| 163 |
+
"step": 100
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.18,
|
| 167 |
+
"learning_rate": 4.9142106826480114e-06,
|
| 168 |
+
"logits/chosen": 0.01861090026795864,
|
| 169 |
+
"logits/rejected": 0.041032858192920685,
|
| 170 |
+
"logps/chosen": -754.0704956054688,
|
| 171 |
+
"logps/rejected": -777.1426391601562,
|
| 172 |
+
"loss": 0.6776,
|
| 173 |
+
"rewards/accuracies": 0.5375000238418579,
|
| 174 |
+
"rewards/chosen": -0.27138790488243103,
|
| 175 |
+
"rewards/margins": 0.0310644693672657,
|
| 176 |
+
"rewards/rejected": -0.3024523854255676,
|
| 177 |
+
"step": 110
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.19,
|
| 181 |
+
"learning_rate": 4.874162703221823e-06,
|
| 182 |
+
"logits/chosen": -0.012570058926939964,
|
| 183 |
+
"logits/rejected": -0.028305906802415848,
|
| 184 |
+
"logps/chosen": -888.3677978515625,
|
| 185 |
+
"logps/rejected": -892.1744384765625,
|
| 186 |
+
"loss": 0.6705,
|
| 187 |
+
"rewards/accuracies": 0.543749988079071,
|
| 188 |
+
"rewards/chosen": -0.34433773159980774,
|
| 189 |
+
"rewards/margins": 0.03458678722381592,
|
| 190 |
+
"rewards/rejected": -0.3789244592189789,
|
| 191 |
+
"step": 120
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.21,
|
| 195 |
+
"learning_rate": 4.826697788369752e-06,
|
| 196 |
+
"logits/chosen": -0.039124295115470886,
|
| 197 |
+
"logits/rejected": 0.005070955958217382,
|
| 198 |
+
"logps/chosen": -856.4949340820312,
|
| 199 |
+
"logps/rejected": -902.6595458984375,
|
| 200 |
+
"loss": 0.6731,
|
| 201 |
+
"rewards/accuracies": 0.581250011920929,
|
| 202 |
+
"rewards/chosen": -0.3688521981239319,
|
| 203 |
+
"rewards/margins": 0.05119430273771286,
|
| 204 |
+
"rewards/rejected": -0.42004650831222534,
|
| 205 |
+
"step": 130
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.22,
|
| 209 |
+
"learning_rate": 4.7719642195082224e-06,
|
| 210 |
+
"logits/chosen": -0.051692504435777664,
|
| 211 |
+
"logits/rejected": 0.06696543842554092,
|
| 212 |
+
"logps/chosen": -899.3610229492188,
|
| 213 |
+
"logps/rejected": -987.4519653320312,
|
| 214 |
+
"loss": 0.6609,
|
| 215 |
+
"rewards/accuracies": 0.6000000238418579,
|
| 216 |
+
"rewards/chosen": -0.38903677463531494,
|
| 217 |
+
"rewards/margins": 0.10688074678182602,
|
| 218 |
+
"rewards/rejected": -0.49591749906539917,
|
| 219 |
+
"step": 140
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.24,
|
| 223 |
+
"learning_rate": 4.710132985485355e-06,
|
| 224 |
+
"logits/chosen": 0.04679523780941963,
|
| 225 |
+
"logits/rejected": 0.16161568462848663,
|
| 226 |
+
"logps/chosen": -1067.924072265625,
|
| 227 |
+
"logps/rejected": -1131.458984375,
|
| 228 |
+
"loss": 0.6745,
|
| 229 |
+
"rewards/accuracies": 0.550000011920929,
|
| 230 |
+
"rewards/chosen": -0.5985130071640015,
|
| 231 |
+
"rewards/margins": 0.07710713148117065,
|
| 232 |
+
"rewards/rejected": -0.6756200790405273,
|
| 233 |
+
"step": 150
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.26,
|
| 237 |
+
"learning_rate": 4.641397248408122e-06,
|
| 238 |
+
"logits/chosen": 0.12509503960609436,
|
| 239 |
+
"logits/rejected": 0.20963266491889954,
|
| 240 |
+
"logps/chosen": -970.3092651367188,
|
| 241 |
+
"logps/rejected": -1117.892333984375,
|
| 242 |
+
"loss": 0.6506,
|
| 243 |
+
"rewards/accuracies": 0.612500011920929,
|
| 244 |
+
"rewards/chosen": -0.4672892987728119,
|
| 245 |
+
"rewards/margins": 0.15404292941093445,
|
| 246 |
+
"rewards/rejected": -0.6213322877883911,
|
| 247 |
+
"step": 160
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.27,
|
| 251 |
+
"learning_rate": 4.5659717401997655e-06,
|
| 252 |
+
"logits/chosen": 0.1800670623779297,
|
| 253 |
+
"logits/rejected": 0.20961984992027283,
|
| 254 |
+
"logps/chosen": -1060.942626953125,
|
| 255 |
+
"logps/rejected": -1264.063232421875,
|
| 256 |
+
"loss": 0.6553,
|
| 257 |
+
"rewards/accuracies": 0.625,
|
| 258 |
+
"rewards/chosen": -0.5244041681289673,
|
| 259 |
+
"rewards/margins": 0.17471647262573242,
|
| 260 |
+
"rewards/rejected": -0.6991206407546997,
|
| 261 |
+
"step": 170
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.29,
|
| 265 |
+
"learning_rate": 4.4840920917726425e-06,
|
| 266 |
+
"logits/chosen": 0.25674110651016235,
|
| 267 |
+
"logits/rejected": 0.2803070843219757,
|
| 268 |
+
"logps/chosen": -1182.87060546875,
|
| 269 |
+
"logps/rejected": -1315.55712890625,
|
| 270 |
+
"loss": 0.6543,
|
| 271 |
+
"rewards/accuracies": 0.5375000238418579,
|
| 272 |
+
"rewards/chosen": -0.7135756611824036,
|
| 273 |
+
"rewards/margins": 0.13786718249320984,
|
| 274 |
+
"rewards/rejected": -0.851442813873291,
|
| 275 |
+
"step": 180
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.3,
|
| 279 |
+
"learning_rate": 4.396014096912182e-06,
|
| 280 |
+
"logits/chosen": 0.1727183610200882,
|
| 281 |
+
"logits/rejected": 0.35109418630599976,
|
| 282 |
+
"logps/chosen": -1045.9290771484375,
|
| 283 |
+
"logps/rejected": -1137.6866455078125,
|
| 284 |
+
"loss": 0.6689,
|
| 285 |
+
"rewards/accuracies": 0.6312500238418579,
|
| 286 |
+
"rewards/chosen": -0.5587235689163208,
|
| 287 |
+
"rewards/margins": 0.10695569217205048,
|
| 288 |
+
"rewards/rejected": -0.6656793355941772,
|
| 289 |
+
"step": 190
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.32,
|
| 293 |
+
"learning_rate": 4.302012913171584e-06,
|
| 294 |
+
"logits/chosen": 0.21685247123241425,
|
| 295 |
+
"logits/rejected": 0.1864197552204132,
|
| 296 |
+
"logps/chosen": -865.9853515625,
|
| 297 |
+
"logps/rejected": -946.7151489257812,
|
| 298 |
+
"loss": 0.6622,
|
| 299 |
+
"rewards/accuracies": 0.6000000238418579,
|
| 300 |
+
"rewards/chosen": -0.3805222511291504,
|
| 301 |
+
"rewards/margins": 0.08247803151607513,
|
| 302 |
+
"rewards/rejected": -0.46300023794174194,
|
| 303 |
+
"step": 200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.34,
|
| 307 |
+
"learning_rate": 4.202382202273702e-06,
|
| 308 |
+
"logits/chosen": 0.3170011639595032,
|
| 309 |
+
"logits/rejected": 0.26561760902404785,
|
| 310 |
+
"logps/chosen": -860.9000854492188,
|
| 311 |
+
"logps/rejected": -995.12353515625,
|
| 312 |
+
"loss": 0.6505,
|
| 313 |
+
"rewards/accuracies": 0.5375000238418579,
|
| 314 |
+
"rewards/chosen": -0.4116226136684418,
|
| 315 |
+
"rewards/margins": 0.10268483310937881,
|
| 316 |
+
"rewards/rejected": -0.5143074989318848,
|
| 317 |
+
"step": 210
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.35,
|
| 321 |
+
"learning_rate": 4.097433212705492e-06,
|
| 322 |
+
"logits/chosen": 0.32917365431785583,
|
| 323 |
+
"logits/rejected": 0.422170490026474,
|
| 324 |
+
"logps/chosen": -1080.322021484375,
|
| 325 |
+
"logps/rejected": -1246.759521484375,
|
| 326 |
+
"loss": 0.6454,
|
| 327 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 328 |
+
"rewards/chosen": -0.6017876863479614,
|
| 329 |
+
"rewards/margins": 0.16769546270370483,
|
| 330 |
+
"rewards/rejected": -0.7694831490516663,
|
| 331 |
+
"step": 220
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.37,
|
| 335 |
+
"learning_rate": 3.987493807371033e-06,
|
| 336 |
+
"logits/chosen": 0.3952299654483795,
|
| 337 |
+
"logits/rejected": 0.39465999603271484,
|
| 338 |
+
"logps/chosen": -1251.5406494140625,
|
| 339 |
+
"logps/rejected": -1388.2440185546875,
|
| 340 |
+
"loss": 0.6442,
|
| 341 |
+
"rewards/accuracies": 0.543749988079071,
|
| 342 |
+
"rewards/chosen": -0.7216378450393677,
|
| 343 |
+
"rewards/margins": 0.1245543584227562,
|
| 344 |
+
"rewards/rejected": -0.8461921811103821,
|
| 345 |
+
"step": 230
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.38,
|
| 349 |
+
"learning_rate": 3.872907439340758e-06,
|
| 350 |
+
"logits/chosen": 0.4063127934932709,
|
| 351 |
+
"logits/rejected": 0.47256404161453247,
|
| 352 |
+
"logps/chosen": -1212.349853515625,
|
| 353 |
+
"logps/rejected": -1397.411865234375,
|
| 354 |
+
"loss": 0.6451,
|
| 355 |
+
"rewards/accuracies": 0.581250011920929,
|
| 356 |
+
"rewards/chosen": -0.7250560522079468,
|
| 357 |
+
"rewards/margins": 0.1564178168773651,
|
| 358 |
+
"rewards/rejected": -0.8814738392829895,
|
| 359 |
+
"step": 240
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.4,
|
| 363 |
+
"learning_rate": 3.75403207889666e-06,
|
| 364 |
+
"logits/chosen": 0.3914504051208496,
|
| 365 |
+
"logits/rejected": 0.3554847240447998,
|
| 366 |
+
"logps/chosen": -1249.439208984375,
|
| 367 |
+
"logps/rejected": -1423.4649658203125,
|
| 368 |
+
"loss": 0.6436,
|
| 369 |
+
"rewards/accuracies": 0.606249988079071,
|
| 370 |
+
"rewards/chosen": -0.7152602076530457,
|
| 371 |
+
"rewards/margins": 0.17244547605514526,
|
| 372 |
+
"rewards/rejected": -0.8877056837081909,
|
| 373 |
+
"step": 250
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.42,
|
| 377 |
+
"learning_rate": 3.631239095225417e-06,
|
| 378 |
+
"logits/chosen": 0.3514712452888489,
|
| 379 |
+
"logits/rejected": 0.45444172620773315,
|
| 380 |
+
"logps/chosen": -1238.7828369140625,
|
| 381 |
+
"logps/rejected": -1391.1258544921875,
|
| 382 |
+
"loss": 0.6492,
|
| 383 |
+
"rewards/accuracies": 0.581250011920929,
|
| 384 |
+
"rewards/chosen": -0.7164724469184875,
|
| 385 |
+
"rewards/margins": 0.16447243094444275,
|
| 386 |
+
"rewards/rejected": -0.8809448480606079,
|
| 387 |
+
"step": 260
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.43,
|
| 391 |
+
"learning_rate": 3.5049120962530608e-06,
|
| 392 |
+
"logits/chosen": 0.375350683927536,
|
| 393 |
+
"logits/rejected": 0.4519611895084381,
|
| 394 |
+
"logps/chosen": -1169.996337890625,
|
| 395 |
+
"logps/rejected": -1291.4034423828125,
|
| 396 |
+
"loss": 0.6529,
|
| 397 |
+
"rewards/accuracies": 0.643750011920929,
|
| 398 |
+
"rewards/chosen": -0.6447538137435913,
|
| 399 |
+
"rewards/margins": 0.12557630240917206,
|
| 400 |
+
"rewards/rejected": -0.7703300714492798,
|
| 401 |
+
"step": 270
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.45,
|
| 405 |
+
"learning_rate": 3.3754457302455464e-06,
|
| 406 |
+
"logits/chosen": 0.3003133237361908,
|
| 407 |
+
"logits/rejected": 0.448310911655426,
|
| 408 |
+
"logps/chosen": -1107.552734375,
|
| 409 |
+
"logps/rejected": -1246.8009033203125,
|
| 410 |
+
"loss": 0.6489,
|
| 411 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 412 |
+
"rewards/chosen": -0.6071010828018188,
|
| 413 |
+
"rewards/margins": 0.13486045598983765,
|
| 414 |
+
"rewards/rejected": -0.7419615387916565,
|
| 415 |
+
"step": 280
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.46,
|
| 419 |
+
"learning_rate": 3.2432444529190714e-06,
|
| 420 |
+
"logits/chosen": 0.3579423427581787,
|
| 421 |
+
"logits/rejected": 0.432018905878067,
|
| 422 |
+
"logps/chosen": -1139.100830078125,
|
| 423 |
+
"logps/rejected": -1201.1663818359375,
|
| 424 |
+
"loss": 0.6495,
|
| 425 |
+
"rewards/accuracies": 0.4749999940395355,
|
| 426 |
+
"rewards/chosen": -0.6468966007232666,
|
| 427 |
+
"rewards/margins": 0.0843476727604866,
|
| 428 |
+
"rewards/rejected": -0.7312443852424622,
|
| 429 |
+
"step": 290
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.48,
|
| 433 |
+
"learning_rate": 3.1087212639117057e-06,
|
| 434 |
+
"logits/chosen": 0.3242015540599823,
|
| 435 |
+
"logits/rejected": 0.4742099642753601,
|
| 436 |
+
"logps/chosen": -1214.9383544921875,
|
| 437 |
+
"logps/rejected": -1312.9287109375,
|
| 438 |
+
"loss": 0.6498,
|
| 439 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 440 |
+
"rewards/chosen": -0.6838893294334412,
|
| 441 |
+
"rewards/margins": 0.11757048219442368,
|
| 442 |
+
"rewards/rejected": -0.801459789276123,
|
| 443 |
+
"step": 300
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.5,
|
| 447 |
+
"learning_rate": 2.9722964165636263e-06,
|
| 448 |
+
"logits/chosen": 0.4161956310272217,
|
| 449 |
+
"logits/rejected": 0.47839951515197754,
|
| 450 |
+
"logps/chosen": -1124.3660888671875,
|
| 451 |
+
"logps/rejected": -1247.5843505859375,
|
| 452 |
+
"loss": 0.6398,
|
| 453 |
+
"rewards/accuracies": 0.543749988079071,
|
| 454 |
+
"rewards/chosen": -0.6321052312850952,
|
| 455 |
+
"rewards/margins": 0.10825280100107193,
|
| 456 |
+
"rewards/rejected": -0.7403579950332642,
|
| 457 |
+
"step": 310
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.51,
|
| 461 |
+
"learning_rate": 2.8343961050366275e-06,
|
| 462 |
+
"logits/chosen": 0.46386805176734924,
|
| 463 |
+
"logits/rejected": 0.5212578773498535,
|
| 464 |
+
"logps/chosen": -1118.0859375,
|
| 465 |
+
"logps/rejected": -1206.055908203125,
|
| 466 |
+
"loss": 0.6473,
|
| 467 |
+
"rewards/accuracies": 0.581250011920929,
|
| 468 |
+
"rewards/chosen": -0.6330564022064209,
|
| 469 |
+
"rewards/margins": 0.102098748087883,
|
| 470 |
+
"rewards/rejected": -0.7351552248001099,
|
| 471 |
+
"step": 320
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.53,
|
| 475 |
+
"learning_rate": 2.695451132874385e-06,
|
| 476 |
+
"logits/chosen": 0.5020041465759277,
|
| 477 |
+
"logits/rejected": 0.4887539744377136,
|
| 478 |
+
"logps/chosen": -1234.7464599609375,
|
| 479 |
+
"logps/rejected": -1376.104736328125,
|
| 480 |
+
"loss": 0.6393,
|
| 481 |
+
"rewards/accuracies": 0.53125,
|
| 482 |
+
"rewards/chosen": -0.7160638570785522,
|
| 483 |
+
"rewards/margins": 0.1427764594554901,
|
| 484 |
+
"rewards/rejected": -0.85884028673172,
|
| 485 |
+
"step": 330
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.54,
|
| 489 |
+
"learning_rate": 2.5558955671628964e-06,
|
| 490 |
+
"logits/chosen": 0.44761380553245544,
|
| 491 |
+
"logits/rejected": 0.5008188486099243,
|
| 492 |
+
"logps/chosen": -1225.663818359375,
|
| 493 |
+
"logps/rejected": -1363.5836181640625,
|
| 494 |
+
"loss": 0.6549,
|
| 495 |
+
"rewards/accuracies": 0.59375,
|
| 496 |
+
"rewards/chosen": -0.7390362620353699,
|
| 497 |
+
"rewards/margins": 0.1303921490907669,
|
| 498 |
+
"rewards/rejected": -0.8694285154342651,
|
| 499 |
+
"step": 340
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.56,
|
| 503 |
+
"learning_rate": 2.4161653824955654e-06,
|
| 504 |
+
"logits/chosen": 0.502762496471405,
|
| 505 |
+
"logits/rejected": 0.521251380443573,
|
| 506 |
+
"logps/chosen": -1289.813720703125,
|
| 507 |
+
"logps/rejected": -1509.394775390625,
|
| 508 |
+
"loss": 0.6337,
|
| 509 |
+
"rewards/accuracies": 0.6499999761581421,
|
| 510 |
+
"rewards/chosen": -0.7973452806472778,
|
| 511 |
+
"rewards/margins": 0.20352289080619812,
|
| 512 |
+
"rewards/rejected": -1.0008680820465088,
|
| 513 |
+
"step": 350
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.58,
|
| 517 |
+
"learning_rate": 2.2766970989791697e-06,
|
| 518 |
+
"logits/chosen": 0.5564379096031189,
|
| 519 |
+
"logits/rejected": 0.4828278422355652,
|
| 520 |
+
"logps/chosen": -1317.9136962890625,
|
| 521 |
+
"logps/rejected": -1490.851806640625,
|
| 522 |
+
"loss": 0.6489,
|
| 523 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 524 |
+
"rewards/chosen": -0.8256872296333313,
|
| 525 |
+
"rewards/margins": 0.14507783949375153,
|
| 526 |
+
"rewards/rejected": -0.9707651138305664,
|
| 527 |
+
"step": 360
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.59,
|
| 531 |
+
"learning_rate": 2.1379264185356545e-06,
|
| 532 |
+
"logits/chosen": 0.4197818338871002,
|
| 533 |
+
"logits/rejected": 0.4475874900817871,
|
| 534 |
+
"logps/chosen": -1211.713623046875,
|
| 535 |
+
"logps/rejected": -1361.810791015625,
|
| 536 |
+
"loss": 0.6479,
|
| 537 |
+
"rewards/accuracies": 0.6000000238418579,
|
| 538 |
+
"rewards/chosen": -0.7186134457588196,
|
| 539 |
+
"rewards/margins": 0.15284962952136993,
|
| 540 |
+
"rewards/rejected": -0.8714631199836731,
|
| 541 |
+
"step": 370
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.61,
|
| 545 |
+
"learning_rate": 2.000286863759934e-06,
|
| 546 |
+
"logits/chosen": 0.4016849398612976,
|
| 547 |
+
"logits/rejected": 0.6099621057510376,
|
| 548 |
+
"logps/chosen": -1292.288818359375,
|
| 549 |
+
"logps/rejected": -1514.7738037109375,
|
| 550 |
+
"loss": 0.6132,
|
| 551 |
+
"rewards/accuracies": 0.612500011920929,
|
| 552 |
+
"rewards/chosen": -0.8104116320610046,
|
| 553 |
+
"rewards/margins": 0.20091453194618225,
|
| 554 |
+
"rewards/rejected": -1.0113261938095093,
|
| 555 |
+
"step": 380
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.62,
|
| 559 |
+
"learning_rate": 1.8642084235859764e-06,
|
| 560 |
+
"logits/chosen": 0.551924467086792,
|
| 561 |
+
"logits/rejected": 0.6596301198005676,
|
| 562 |
+
"logps/chosen": -1387.1436767578125,
|
| 563 |
+
"logps/rejected": -1532.2877197265625,
|
| 564 |
+
"loss": 0.6588,
|
| 565 |
+
"rewards/accuracies": 0.5562499761581421,
|
| 566 |
+
"rewards/chosen": -0.913176417350769,
|
| 567 |
+
"rewards/margins": 0.1407497525215149,
|
| 568 |
+
"rewards/rejected": -1.0539262294769287,
|
| 569 |
+
"step": 390
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.64,
|
| 573 |
+
"learning_rate": 1.7301162099921013e-06,
|
| 574 |
+
"logits/chosen": 0.5529540777206421,
|
| 575 |
+
"logits/rejected": 0.5772652626037598,
|
| 576 |
+
"logps/chosen": -1380.193115234375,
|
| 577 |
+
"logps/rejected": -1582.646484375,
|
| 578 |
+
"loss": 0.6341,
|
| 579 |
+
"rewards/accuracies": 0.5375000238418579,
|
| 580 |
+
"rewards/chosen": -0.8533477783203125,
|
| 581 |
+
"rewards/margins": 0.18637995421886444,
|
| 582 |
+
"rewards/rejected": -1.0397276878356934,
|
| 583 |
+
"step": 400
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.66,
|
| 587 |
+
"learning_rate": 1.5984291299420117e-06,
|
| 588 |
+
"logits/chosen": 0.47877341508865356,
|
| 589 |
+
"logits/rejected": 0.5813654661178589,
|
| 590 |
+
"logps/chosen": -1256.6849365234375,
|
| 591 |
+
"logps/rejected": -1472.9185791015625,
|
| 592 |
+
"loss": 0.6404,
|
| 593 |
+
"rewards/accuracies": 0.5625,
|
| 594 |
+
"rewards/chosen": -0.7804775238037109,
|
| 595 |
+
"rewards/margins": 0.19573338329792023,
|
| 596 |
+
"rewards/rejected": -0.97621089220047,
|
| 597 |
+
"step": 410
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.67,
|
| 601 |
+
"learning_rate": 1.4695585767104092e-06,
|
| 602 |
+
"logits/chosen": 0.41678476333618164,
|
| 603 |
+
"logits/rejected": 0.6133105754852295,
|
| 604 |
+
"logps/chosen": -1238.5712890625,
|
| 605 |
+
"logps/rejected": -1406.7432861328125,
|
| 606 |
+
"loss": 0.6409,
|
| 607 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 608 |
+
"rewards/chosen": -0.7545350193977356,
|
| 609 |
+
"rewards/margins": 0.1725122034549713,
|
| 610 |
+
"rewards/rejected": -0.9270472526550293,
|
| 611 |
+
"step": 420
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.69,
|
| 615 |
+
"learning_rate": 1.3439071446815452e-06,
|
| 616 |
+
"logits/chosen": 0.4347582459449768,
|
| 617 |
+
"logits/rejected": 0.5867087244987488,
|
| 618 |
+
"logps/chosen": -1281.6087646484375,
|
| 619 |
+
"logps/rejected": -1513.9439697265625,
|
| 620 |
+
"loss": 0.6387,
|
| 621 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 622 |
+
"rewards/chosen": -0.7874932289123535,
|
| 623 |
+
"rewards/margins": 0.2120811492204666,
|
| 624 |
+
"rewards/rejected": -0.9995743632316589,
|
| 625 |
+
"step": 430
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.7,
|
| 629 |
+
"learning_rate": 1.2218673716356919e-06,
|
| 630 |
+
"logits/chosen": 0.5029743909835815,
|
| 631 |
+
"logits/rejected": 0.591041088104248,
|
| 632 |
+
"logps/chosen": -1364.179931640625,
|
| 633 |
+
"logps/rejected": -1468.728759765625,
|
| 634 |
+
"loss": 0.6623,
|
| 635 |
+
"rewards/accuracies": 0.59375,
|
| 636 |
+
"rewards/chosen": -0.8501272201538086,
|
| 637 |
+
"rewards/margins": 0.09424228221178055,
|
| 638 |
+
"rewards/rejected": -0.9443694949150085,
|
| 639 |
+
"step": 440
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.72,
|
| 643 |
+
"learning_rate": 1.103820512452661e-06,
|
| 644 |
+
"logits/chosen": 0.53369140625,
|
| 645 |
+
"logits/rejected": 0.5491160154342651,
|
| 646 |
+
"logps/chosen": -1355.01171875,
|
| 647 |
+
"logps/rejected": -1442.1351318359375,
|
| 648 |
+
"loss": 0.6686,
|
| 649 |
+
"rewards/accuracies": 0.5562499761581421,
|
| 650 |
+
"rewards/chosen": -0.8647180795669556,
|
| 651 |
+
"rewards/margins": 0.09323947131633759,
|
| 652 |
+
"rewards/rejected": -0.9579576253890991,
|
| 653 |
+
"step": 450
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.74,
|
| 657 |
+
"learning_rate": 9.901353480633468e-07,
|
| 658 |
+
"logits/chosen": 0.5199618339538574,
|
| 659 |
+
"logits/rejected": 0.4953738749027252,
|
| 660 |
+
"logps/chosen": -1270.599853515625,
|
| 661 |
+
"logps/rejected": -1456.319580078125,
|
| 662 |
+
"loss": 0.6241,
|
| 663 |
+
"rewards/accuracies": 0.612500011920929,
|
| 664 |
+
"rewards/chosen": -0.7807380557060242,
|
| 665 |
+
"rewards/margins": 0.18214623630046844,
|
| 666 |
+
"rewards/rejected": -0.9628843069076538,
|
| 667 |
+
"step": 460
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.75,
|
| 671 |
+
"learning_rate": 8.811670333701544e-07,
|
| 672 |
+
"logits/chosen": 0.5168347954750061,
|
| 673 |
+
"logits/rejected": 0.6290295124053955,
|
| 674 |
+
"logps/chosen": -1292.0872802734375,
|
| 675 |
+
"logps/rejected": -1423.704345703125,
|
| 676 |
+
"loss": 0.639,
|
| 677 |
+
"rewards/accuracies": 0.59375,
|
| 678 |
+
"rewards/chosen": -0.771618664264679,
|
| 679 |
+
"rewards/margins": 0.1294311136007309,
|
| 680 |
+
"rewards/rejected": -0.901049792766571,
|
| 681 |
+
"step": 470
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.77,
|
| 685 |
+
"learning_rate": 7.772559877354341e-07,
|
| 686 |
+
"logits/chosen": 0.5879518985748291,
|
| 687 |
+
"logits/rejected": 0.6285650730133057,
|
| 688 |
+
"logps/chosen": -1255.3343505859375,
|
| 689 |
+
"logps/rejected": -1419.4129638671875,
|
| 690 |
+
"loss": 0.6316,
|
| 691 |
+
"rewards/accuracies": 0.574999988079071,
|
| 692 |
+
"rewards/chosen": -0.8007827997207642,
|
| 693 |
+
"rewards/margins": 0.14159244298934937,
|
| 694 |
+
"rewards/rejected": -0.9423751831054688,
|
| 695 |
+
"step": 480
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.78,
|
| 699 |
+
"learning_rate": 6.787268315040604e-07,
|
| 700 |
+
"logits/chosen": 0.45197635889053345,
|
| 701 |
+
"logits/rejected": 0.5737031698226929,
|
| 702 |
+
"logps/chosen": -1276.957275390625,
|
| 703 |
+
"logps/rejected": -1509.0966796875,
|
| 704 |
+
"loss": 0.6336,
|
| 705 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 706 |
+
"rewards/chosen": -0.8125177621841431,
|
| 707 |
+
"rewards/margins": 0.19976083934307098,
|
| 708 |
+
"rewards/rejected": -1.012278437614441,
|
| 709 |
+
"step": 490
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.8,
|
| 713 |
+
"learning_rate": 5.858873718824829e-07,
|
| 714 |
+
"logits/chosen": 0.5449025630950928,
|
| 715 |
+
"logits/rejected": 0.6064163446426392,
|
| 716 |
+
"logps/chosen": -1345.1566162109375,
|
| 717 |
+
"logps/rejected": -1465.4927978515625,
|
| 718 |
+
"loss": 0.6838,
|
| 719 |
+
"rewards/accuracies": 0.5562499761581421,
|
| 720 |
+
"rewards/chosen": -0.8671814799308777,
|
| 721 |
+
"rewards/margins": 0.12205106019973755,
|
| 722 |
+
"rewards/rejected": -0.9892325401306152,
|
| 723 |
+
"step": 500
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.82,
|
| 727 |
+
"learning_rate": 4.990276413423817e-07,
|
| 728 |
+
"logits/chosen": 0.45466941595077515,
|
| 729 |
+
"logits/rejected": 0.5977617502212524,
|
| 730 |
+
"logps/chosen": -1280.716064453125,
|
| 731 |
+
"logps/rejected": -1477.654541015625,
|
| 732 |
+
"loss": 0.6267,
|
| 733 |
+
"rewards/accuracies": 0.612500011920929,
|
| 734 |
+
"rewards/chosen": -0.7988293766975403,
|
| 735 |
+
"rewards/margins": 0.19528616964817047,
|
| 736 |
+
"rewards/rejected": -0.9941156506538391,
|
| 737 |
+
"step": 510
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.83,
|
| 741 |
+
"learning_rate": 4.184189915529796e-07,
|
| 742 |
+
"logits/chosen": 0.5602769255638123,
|
| 743 |
+
"logits/rejected": 0.611350953578949,
|
| 744 |
+
"logps/chosen": -1306.686279296875,
|
| 745 |
+
"logps/rejected": -1565.746826171875,
|
| 746 |
+
"loss": 0.6323,
|
| 747 |
+
"rewards/accuracies": 0.675000011920929,
|
| 748 |
+
"rewards/chosen": -0.794352650642395,
|
| 749 |
+
"rewards/margins": 0.22704383730888367,
|
| 750 |
+
"rewards/rejected": -1.0213963985443115,
|
| 751 |
+
"step": 520
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.85,
|
| 755 |
+
"learning_rate": 3.4431324567258176e-07,
|
| 756 |
+
"logits/chosen": 0.5258222222328186,
|
| 757 |
+
"logits/rejected": 0.5747347474098206,
|
| 758 |
+
"logps/chosen": -1296.38916015625,
|
| 759 |
+
"logps/rejected": -1511.482666015625,
|
| 760 |
+
"loss": 0.6341,
|
| 761 |
+
"rewards/accuracies": 0.606249988079071,
|
| 762 |
+
"rewards/chosen": -0.7807685136795044,
|
| 763 |
+
"rewards/margins": 0.19532978534698486,
|
| 764 |
+
"rewards/rejected": -0.9760982394218445,
|
| 765 |
+
"step": 530
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.86,
|
| 769 |
+
"learning_rate": 2.769419116476052e-07,
|
| 770 |
+
"logits/chosen": 0.46619558334350586,
|
| 771 |
+
"logits/rejected": 0.6566539406776428,
|
| 772 |
+
"logps/chosen": -1344.41748046875,
|
| 773 |
+
"logps/rejected": -1466.8358154296875,
|
| 774 |
+
"loss": 0.6556,
|
| 775 |
+
"rewards/accuracies": 0.59375,
|
| 776 |
+
"rewards/chosen": -0.8279752731323242,
|
| 777 |
+
"rewards/margins": 0.1466597020626068,
|
| 778 |
+
"rewards/rejected": -0.9746350049972534,
|
| 779 |
+
"step": 540
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.88,
|
| 783 |
+
"learning_rate": 2.1651545897676512e-07,
|
| 784 |
+
"logits/chosen": 0.5445398092269897,
|
| 785 |
+
"logits/rejected": 0.5394152998924255,
|
| 786 |
+
"logps/chosen": -1226.078125,
|
| 787 |
+
"logps/rejected": -1361.6815185546875,
|
| 788 |
+
"loss": 0.6609,
|
| 789 |
+
"rewards/accuracies": 0.4937500059604645,
|
| 790 |
+
"rewards/chosen": -0.7584909200668335,
|
| 791 |
+
"rewards/margins": 0.11591245234012604,
|
| 792 |
+
"rewards/rejected": -0.8744033575057983,
|
| 793 |
+
"step": 550
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.9,
|
| 797 |
+
"learning_rate": 1.6322266119983222e-07,
|
| 798 |
+
"logits/chosen": 0.5224272012710571,
|
| 799 |
+
"logits/rejected": 0.6026440858840942,
|
| 800 |
+
"logps/chosen": -1239.085693359375,
|
| 801 |
+
"logps/rejected": -1359.008544921875,
|
| 802 |
+
"loss": 0.6537,
|
| 803 |
+
"rewards/accuracies": 0.550000011920929,
|
| 804 |
+
"rewards/chosen": -0.7636959552764893,
|
| 805 |
+
"rewards/margins": 0.1270333230495453,
|
| 806 |
+
"rewards/rejected": -0.8907292485237122,
|
| 807 |
+
"step": 560
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.91,
|
| 811 |
+
"learning_rate": 1.1723000616502167e-07,
|
| 812 |
+
"logits/chosen": 0.44439181685447693,
|
| 813 |
+
"logits/rejected": 0.587325930595398,
|
| 814 |
+
"logps/chosen": -1246.112548828125,
|
| 815 |
+
"logps/rejected": -1382.4180908203125,
|
| 816 |
+
"loss": 0.6502,
|
| 817 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 818 |
+
"rewards/chosen": -0.7664066553115845,
|
| 819 |
+
"rewards/margins": 0.1451866328716278,
|
| 820 |
+
"rewards/rejected": -0.9115932583808899,
|
| 821 |
+
"step": 570
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.93,
|
| 825 |
+
"learning_rate": 7.868117591737585e-08,
|
| 826 |
+
"logits/chosen": 0.5547848343849182,
|
| 827 |
+
"logits/rejected": 0.6335003972053528,
|
| 828 |
+
"logps/chosen": -1389.32470703125,
|
| 829 |
+
"logps/rejected": -1530.272705078125,
|
| 830 |
+
"loss": 0.6396,
|
| 831 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 832 |
+
"rewards/chosen": -0.8785858154296875,
|
| 833 |
+
"rewards/margins": 0.14252634346485138,
|
| 834 |
+
"rewards/rejected": -1.0211122035980225,
|
| 835 |
+
"step": 580
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.94,
|
| 839 |
+
"learning_rate": 4.769659783295383e-08,
|
| 840 |
+
"logits/chosen": 0.5736573338508606,
|
| 841 |
+
"logits/rejected": 0.6307505369186401,
|
| 842 |
+
"logps/chosen": -1240.878662109375,
|
| 843 |
+
"logps/rejected": -1390.9312744140625,
|
| 844 |
+
"loss": 0.6415,
|
| 845 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 846 |
+
"rewards/chosen": -0.8237506747245789,
|
| 847 |
+
"rewards/margins": 0.1187644749879837,
|
| 848 |
+
"rewards/rejected": -0.9425151944160461,
|
| 849 |
+
"step": 590
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.96,
|
| 853 |
+
"learning_rate": 2.4373068401120358e-08,
|
| 854 |
+
"logits/chosen": 0.4911600947380066,
|
| 855 |
+
"logits/rejected": 0.6507722735404968,
|
| 856 |
+
"logps/chosen": -1280.173095703125,
|
| 857 |
+
"logps/rejected": -1408.863037109375,
|
| 858 |
+
"loss": 0.6423,
|
| 859 |
+
"rewards/accuracies": 0.5562499761581421,
|
| 860 |
+
"rewards/chosen": -0.7916342616081238,
|
| 861 |
+
"rewards/margins": 0.13446274399757385,
|
| 862 |
+
"rewards/rejected": -0.92609703540802,
|
| 863 |
+
"step": 600
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.98,
|
| 867 |
+
"learning_rate": 8.78345083022425e-09,
|
| 868 |
+
"logits/chosen": 0.438457727432251,
|
| 869 |
+
"logits/rejected": 0.5983772277832031,
|
| 870 |
+
"logps/chosen": -1361.9815673828125,
|
| 871 |
+
"logps/rejected": -1418.54541015625,
|
| 872 |
+
"loss": 0.6503,
|
| 873 |
+
"rewards/accuracies": 0.4937500059604645,
|
| 874 |
+
"rewards/chosen": -0.8475711941719055,
|
| 875 |
+
"rewards/margins": 0.06368283927440643,
|
| 876 |
+
"rewards/rejected": -0.9112540483474731,
|
| 877 |
+
"step": 610
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.99,
|
| 881 |
+
"learning_rate": 9.764474213677654e-10,
|
| 882 |
+
"logits/chosen": 0.4755234122276306,
|
| 883 |
+
"logits/rejected": 0.6281898617744446,
|
| 884 |
+
"logps/chosen": -1357.261962890625,
|
| 885 |
+
"logps/rejected": -1532.4453125,
|
| 886 |
+
"loss": 0.6579,
|
| 887 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 888 |
+
"rewards/chosen": -0.8383301496505737,
|
| 889 |
+
"rewards/margins": 0.17498043179512024,
|
| 890 |
+
"rewards/rejected": -1.0133105516433716,
|
| 891 |
+
"step": 620
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.0,
|
| 895 |
+
"step": 625,
|
| 896 |
+
"total_flos": 0.0,
|
| 897 |
+
"train_loss": 0.6552388866424561,
|
| 898 |
+
"train_runtime": 7636.4902,
|
| 899 |
+
"train_samples_per_second": 2.619,
|
| 900 |
+
"train_steps_per_second": 0.082
|
| 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 |
+
}
|