---
library_name: transformers
license: llama3.2
base_model: meta-llama/Llama-3.2-3B
tags:
- generated_from_trainer
datasets:
- ptllama/acemath_test
model-index:
- name: outputs/out
  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. -->

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

axolotl version: `0.7.0`
```yaml
base_model: meta-llama/Llama-3.2-3B
# Automatically upload checkpoint and final model to HF
# hub_model_id: username/custom_model_name

load_in_8bit: false
load_in_4bit: false
strict: false

datasets:
  - path: ptllama/acemath_test
    type: completion

# pretraining_dataset:
#   - name:
#     path: ptllama/acemath_test
#     split:
#     text_column: text # column in dataset with the data, usually `text`
#     type: pretrain
#     trust_remote_code:
#     skip: # number of rows of data to skip over from the beginning

dataset_prepared_path: last_run_prepared
val_set_size: 0.01
output_dir: ./outputs/out

sequence_len: 4096
sample_packing: true
pad_to_sequence_len: true
eval_sample_packing: false

wandb_project: axolotl-pretraining
wandb_entity:
wandb_watch:
wandb_name: test-2e4
wandb_log_model:

gradient_accumulation_steps: 16
micro_batch_size: 4
num_epochs: 1
optimizer: paged_adamw_8bit
lr_scheduler: cosine
learning_rate: 2e-4

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

gradient_checkpointing: true
gradient_checkpointing_kwargs:
  use_reentrant: false
early_stopping_patience:
resume_from_checkpoint:
logging_steps: 1
xformers_attention:
flash_attention: true

warmup_ratio: 0.01
cosine_min_lr_ratio: 0.1
cosine_constant_lr_ratio: 0.9
evals_per_epoch: 2
eval_table_size:
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:
  pad_token: <|end_of_text|>

```

</details><br>

# outputs/out

This model is a fine-tuned version of [meta-llama/Llama-3.2-3B](https://huggingface.co/meta-llama/Llama-3.2-3B) on the ptllama/acemath_test dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2855

## 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: 8
- gradient_accumulation_steps: 16
- total_train_batch_size: 512
- total_eval_batch_size: 32
- optimizer: Use paged_adamw_8bit with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 17
- num_epochs: 1.0

### Training results

| Training Loss | Epoch  | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 1.1349        | 0.0006 | 1    | 1.2094          |
| 0.2805        | 0.5002 | 893  | 0.2855          |


### Framework versions

- Transformers 4.48.3
- Pytorch 2.4.0
- Datasets 3.2.0
- Tokenizers 0.21.0