AlekseyKorshuk/pythia-1b-deduped-chatml

<!-- 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. -->

pythia-1b-deduped-chatml

This model is a fine-tuned version of EleutherAI/pythia-1b-deduped on the None dataset. It achieves the following results on the evaluation set:

• —Loss: 1.0088
• —Accuracy: 0.2334
• —Entropy: 0.8955

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: 2
• —gradientaccumulationsteps: 2
• —totaltrainbatch_size: 32
• —totalevalbatch_size: 16
• —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• —lrschedulertype: cosine
• —lrschedulerwarmup_ratio: 0.03
• —num_epochs: 2

Training results

Framework versions

• —Transformers 4.28.1
• —Pytorch 2.0.0-rc1
• —Datasets 2.10.1
• —Tokenizers 0.13.3