---
title: seed_math_deepmind
canonical_url: "https://www.modelscope.cn/models/mlfoundations-dev/seed_math_deepmind"
md_url: "https://www.modelscope.cn/models/mlfoundations-dev/seed_math_deepmind.md"
repository: mlfoundations-dev/seed_math_deepmind
last_updated: 2025-09-28
license: llama3.1
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - llama
architectures:
  - LlamaForCausalLM
base_model:
  - meta-llama/Meta-Llama-3.1-8B
base_model_relation: finetune
parameters: 8.0B
tensor_type:
  - BF16
library_name:
  - safetensors
  - pytorch
frameworks:
  - pytorch
inference_backends:
  - "deploy_task text/emb"
  - "lmdeploy 0.9.1"
  - "lmdeploy_turbomind 0.9.1"
  - "sglang 0.5.2"
  - "vllm 0.9.2"
downloads: 16
stars: 0
tags:
  - llama-factory
  - full
  - generated_from_trainer
---

# seed_math_deepmind

> seed_math_deepmind - mlfoundations-dev 在 ModelScope 开源的模型。This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B on the mlfoundations-dev/seedmathdeepmind dataset. It achieves the following results on the evaluation set: Loss: 0.1540

mlfoundations-dev/seed_math_deepmind 是 ModelScope 魔搭社区上的 8.0B 参数text-generation模型，采用 llama3.1 许可，基于 meta-llama/Meta-Llama-3.1-8B 构建，可用 deploy_task text/emb、lmdeploy 0.9.1、lmdeploy_turbomind 0.9.1 部署。

- **Repository**: mlfoundations-dev/seed_math_deepmind
- **License**: llama3.1
- **Tasks**: text-generation
- **Parameters**: 8.0B
- **Base model**: meta-llama/Meta-Llama-3.1-8B
- **Inference backends**: deploy_task text/emb, lmdeploy 0.9.1, lmdeploy_turbomind 0.9.1, sglang 0.5.2, vllm 0.9.2
- **Tags**: llama-factory, full, generated_from_trainer
- **Downloads**: 16
- **Stars**: 0
- **Last updated**: 2025-09-28

Source: https://www.modelscope.cn/models/mlfoundations-dev/seed_math_deepmind

---

<!-- 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. -->

# seed_math_deepmind

This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the mlfoundations-dev/seed_math_deepmind dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1540

## 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: 5e-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 8
- total_train_batch_size: 512
- total_eval_batch_size: 64
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: constant
- num_epochs: 3.0

### Training results

| Training Loss | Epoch  | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 0.2081        | 0.9730 | 27   | 0.1813          |
| 0.1642        | 1.9685 | 54   | 0.1608          |
| 0.1506        | 2.9640 | 81   | 0.1540          |


### Framework versions

- Transformers 4.46.1
- Pytorch 2.3.0
- Datasets 3.1.0
- Tokenizers 0.20.3