IeBoytsov/llama3-1-ox-llms-8b-sft-full-germany-data

<!-- 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. -->

llama3-1-ox-llms-8b-sft-full-germany-data

This model is a fine-tuned version of meta-llama/Llama-3.1-8B on the oxford-llms/ultrachatfiltered, the oxford-llms/Magpie-Qwen2.5-Pro-1M-v0.1-filtered, the oxford-llms/europeansocialsurvey2020sft, the oxford-llms/europeansocialsurvey2023sft, the oxford-llms/worldvaluessurvey20172022sft and the oxford-llms/europeansocialsurvey2023germany_sft datasets. It achieves the following results on the evaluation set:

• —Loss: 0.7417

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: 2e-05
• —trainbatchsize: 8
• —evalbatchsize: 8
• —seed: 42
• —distributed_type: multi-GPU
• —num_devices: 8
• —totaltrainbatch_size: 64
• —totalevalbatch_size: 64
• —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• —lrschedulertype: cosine
• —lrschedulerwarmup_ratio: 0.1
• —num_epochs: 2

Training results

Framework versions

• —Transformers 4.45.1
• —Pytorch 2.5.1+cu124
• —Datasets 3.1.0
• —Tokenizers 0.20.3