---
library_name: peft
license: mit
base_model: microsoft/Phi-3.5-mini-instruct
tags:
- axolotl
- generated_from_trainer
model-index:
- name: sn_math_curator_on_ensemble_8
  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. -->


This is an open-source fine-tuned reasoning adapter of [microsoft/Phi-3.5-mini-instruct](https://huggingface.co/microsoft/Phi-3.5-mini-instruct), transformed into a math reasoning model using data curated from [collinear-ai/R1-Distill-SFT-Curated](https://huggingface.co/datasets/collinear-ai/R1-Distill-SFT-Curated).

[<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 version</summary>

axolotl version: `0.5.0`
<!-- ```yaml
strict: false
base_model: microsoft/Phi-3.5-mini-instruct
tokenizer_config: microsoft/Phi-3.5-mini-instruct
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer

# Output configuration
hub_model_id: collinear-ai/sn_math_curator_on_ensemble_8
dataset_prepared_path: data/sn_math_curator_on_ensemble_8
output_dir: model/sn_math_curator_on_ensemble_8

# Format the dataset into the right instruction format.
chat_template: phi_3
datasets:
  - path: collinear-ai/R1-Distill-SFT-numina-math-ensemble_8_train
    split: train
    type: chat_template
    chat_template: phi_3
    field_messages: train_conv
    message_field_role: role
    message_field_content: content
train_on_inputs: false #FALSE

val_set_size: 0.05
# Data packing
sequence_len: 4096
eval_sample_packing: false
sample_packing: false
pad_to_sequence_len: true
group_by_length: false

# Lora config
adapter: qlora
lora_model_dir:
load_in_8bit: false
load_in_4bit: true -->
<!-- lora_r: 128
lora_alpha: 64
lora_dropout: 0.2
lora_target_linear: true
lora_fan_in_fan_out:
lora_target_modules:
  - gate_proj
  - down_proj
  - up_proj
  - q_proj
  - v_proj
  - k_proj
  - o_proj
lora_modules_to_save: 
  - embed_tokens
  - lm_head


# Logging config
wandb_project: sn-curators-downstream
wandb_entity: nazneen
wandb_name: curator_math_sn_ensemble_8_phi

# Trainer config
gradient_accumulation_steps: 2
micro_batch_size: 10
num_epochs: 1
optimizer: paged_adamw_8bit
lr_scheduler: cosine
learning_rate: 0.000005


bfloat16: true
bf16: true
fp16:
tf32: false -->

<!-- gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 10
xformers_attention:
flash_attention: true
save_safetensors: true



warmup_steps: 50
evals_per_epoch: 3
eval_table_size: 3
eval_max_new_tokens: 2048
saves_per_epoch: 40
debug:
deepspeed:
weight_decay: 0.02
fsdp_config:
special_tokens:
  bos_token: "<s>"
  eos_token: "<|endoftext|>"
  unk_token: "<unk>"
  pad_token: "<|endoftext|>"

``` -->

</details><br>



## Intended uses & limitations

Math-Reasoning

## Training and evaluation data

Training data curated from [collinear-ai/R1-Distill-SFT-Curated](https://huggingface.co/datasets/collinear-ai/R1-Distill-SFT-Curated)
Evaluation data: [HuggingFaceH4/MATH-500](https://huggingface.co/datasets/HuggingFaceH4/MATH-500)

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 5e-06
- train_batch_size: 10
- eval_batch_size: 10
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 2
- total_train_batch_size: 160
- total_eval_batch_size: 80
- optimizer: Use OptimizerNames.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: 50
- num_epochs: 1

### Training results

| Training Loss | Epoch  | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| No log        | 0.0003 | 1    | 0.6646          |
| 0.3174        | 0.3335 | 1247 | 0.3329          |
| 0.307         | 0.6670 | 2494 | 0.3169          |

### Evaluation on Math500

![Math Reasoning Evaluation](https://huggingface.co/collinear-ai/math_reasoning_phi_c2/raw/main/math500_eval_c1_c2.png)


### Framework versions

- PEFT 0.13.2
- Transformers 4.46.1
- Pytorch 2.3.1+cu121
- Datasets 3.0.1
- Tokenizers 0.20.3