---
license: other
library_name: peft
tags:
- generated_from_trainer
base_model: google/gemma-7b-it
model-index:
- name: out
  results: []
pipeline_tag: text-generation
---

<!-- 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. -->

[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
<details><summary>See axolotl config</summary>

axolotl version: `0.4.0`
```yaml
# use google/gemma-7b if you have access
base_model: google/gemma-7b-it
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer

load_in_8bit: false
load_in_4bit: true
strict: false

# huggingface repo
datasets:
  - path: ./python-oasst/chunk_1.jsonl
    type: oasst
val_set_size: 0.1
output_dir: ./out

adapter: qlora
lora_r: 32
lora_alpha: 16
lora_dropout: 0.05
lora_target_linear: true

sequence_len: 4096
sample_packing: false
pad_to_sequence_len: true

wandb_project: gemma-7b-it
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:


gradient_accumulation_steps: 6
micro_batch_size: 4
num_epochs: 4
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.0002

train_on_inputs: true
group_by_length: false
bf16: auto
fp16:
tf32: false

gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true

warmup_ratio: 0.1
evals_per_epoch: 4
eval_table_size:
eval_max_new_tokens: 128
saves_per_epoch: 1
debug:
deepspeed: deepspeed_configs/zero1.json
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:

```

</details><br>

# out

This model is a fine-tuned version of [google/gemma-7b-it](https://huggingface.co/google/gemma-7b-it) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.1905

## 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: 0.0002
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 6
- total_train_batch_size: 96
- total_eval_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 9
- num_epochs: 4

### Training results

| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 5.0474        | 0.01  | 1    | 5.9279          |
| 1.221         | 0.26  | 24   | 1.2960          |
| 1.1167        | 0.51  | 48   | 1.1657          |
| 1.0702        | 0.77  | 72   | 1.1372          |
| 0.9553        | 1.02  | 96   | 1.1292          |
| 0.9294        | 1.28  | 120  | 1.1301          |
| 0.9603        | 1.54  | 144  | 1.1254          |
| 0.8544        | 1.79  | 168  | 1.1276          |
| 0.826         | 2.05  | 192  | 1.1462          |
| 0.816         | 2.31  | 216  | 1.1500          |
| 0.7392        | 2.56  | 240  | 1.1446          |
| 0.7597        | 2.82  | 264  | 1.1469          |
| 0.6664        | 3.07  | 288  | 1.1908          |
| 0.6968        | 3.33  | 312  | 1.1842          |
| 0.7327        | 3.59  | 336  | 1.1899          |
| 0.7211        | 3.84  | 360  | 1.1905          |


### Framework versions

- PEFT 0.9.0
- Transformers 4.39.0.dev0
- Pytorch 2.1.2+cu118
- Datasets 2.18.0
- Tokenizers 0.15.0