---
base_model: google/gemma-7b
datasets:
- llama-duo/synth_summarize_dataset_dedup
library_name: peft
license: gemma
tags:
- alignment-handbook
- trl
- sft
- generated_from_trainer
model-index:
- name: gemma7b-gpt4o_1k_summarize-ksaslora-wo-auxloss
  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. -->

# gemma7b-gpt4o_1k_summarize-ksaslora-wo-auxloss

This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the llama-duo/synth_summarize_dataset_dedup dataset.
It achieves the following results on the evaluation set:
- Loss: 2.4203

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure


The following `bitsandbytes` quantization config was used during training:
- quant_method: QuantizationMethod.BITS_AND_BYTES
- _load_in_8bit: True
- _load_in_4bit: False
- llm_int8_threshold: 6.0
- llm_int8_skip_modules: None
- llm_int8_enable_fp32_cpu_offload: False
- llm_int8_has_fp16_weight: False
- bnb_4bit_quant_type: fp4
- bnb_4bit_use_double_quant: False
- bnb_4bit_compute_dtype: float32
- bnb_4bit_quant_storage: uint8
- load_in_4bit: False
- load_in_8bit: True

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 0.0003
- train_batch_size: 1
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 6
- gradient_accumulation_steps: 2
- total_train_batch_size: 12
- total_eval_batch_size: 12
- 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 |
|:-------------:|:-----:|:----:|:---------------:|
| 0.8382        | 1.0   | 1045 | 2.4203          |


### Framework versions

- PEFT 0.6.3.dev0
- Transformers 4.45.0
- Pytorch 2.4.1+cu121
- Datasets 3.0.0
- Tokenizers 0.20.0