File size: 2,037 Bytes
60b088a a5481d3 60b088a a5481d3 60b088a 8abbf99 60b088a 8abbf99 60b088a d7c9bb7 a5481d3 60b088a |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 |
---
base_model: meta-llama/Meta-Llama-3-8B-Instruct
library_name: peft
license: llama3
tags:
- trl
- kto
- generated_from_trainer
model-index:
- name: llama3_false_positives_0609_KTO_hp_screening_seeds
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# llama3_false_positives_0609_KTO_hp_screening_seeds
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.7798
- Eval/rewards/chosen: 0.5857
- Eval/logps/chosen: -186.9679
- Eval/rewards/rejected: 0.0902
- Eval/logps/rejected: -228.9668
- Eval/rewards/margins: 0.4955
- Eval/kl: 5.4514
## 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: 0.0001
- train_batch_size: 1
- eval_batch_size: 2
- seed: 1234
- gradient_accumulation_steps: 8
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 6.0
### Training results
| Training Loss | Epoch | Step | Validation Loss | |
|:-------------:|:-----:|:----:|:---------------:|:------:|
| 0.4717 | 0.96 | 12 | 0.7882 | 0.0 |
| 0.6064 | 2.0 | 25 | 0.8020 | 0.0 |
| 0.2324 | 2.96 | 37 | 0.7275 | 0.0 |
| 0.4044 | 4.0 | 50 | 0.7863 | 3.8025 |
| 0.1764 | 4.96 | 62 | 0.7797 | 5.4128 |
| 0.1937 | 5.76 | 72 | 0.7798 | 5.4514 |
### Framework versions
- PEFT 0.11.1
- Transformers 4.44.0
- Pytorch 2.2.0
- Datasets 2.20.0
- Tokenizers 0.19.1 |