metadata
library_name: transformers
license: llama3
base_model: meta-llama/Meta-Llama-3-8B-Instruct
tags:
- alignment-handbook
- generated_from_trainer
datasets:
- princeton-nlp/llama3-ultrafeedback-armorm
model-index:
- name: llama-3-8b-instruct-gapo-v2-bleu-beta10-gamma0.3-lr1.0e-6-he_scale-rerun
results: []
llama-3-8b-instruct-gapo-v2-bleu-beta10-gamma0.3-lr1.0e-6-he_scale-rerun
This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B-Instruct on the princeton-nlp/llama3-ultrafeedback-armorm dataset. It achieves the following results on the evaluation set:
- Loss: 1.3686
- Rewards/chosen: -14.9003
- Rewards/rejected: -20.1221
- Rewards/accuracies: 0.8354
- Rewards/margins: 5.2218
- Logps/rejected: -2.0122
- Logps/chosen: -1.4900
- Logits/rejected: -1.4927
- Logits/chosen: -1.4845
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-06
- train_batch_size: 2
- eval_batch_size: 4
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 16
- total_train_batch_size: 128
- total_eval_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1.4439 | 0.8550 | 400 | 1.3686 | -14.9003 | -20.1221 | 0.8354 | 5.2218 | -2.0122 | -1.4900 | -1.4927 | -1.4845 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.2.0
- Datasets 2.21.0
- Tokenizers 0.19.1