pythia-160m-tulu-v2-mix

This model is a fine-tuned version of EleutherAI/pythia-160m on the allenai/tulu-v2-sft-mixture dataset. It achieves the following results on the evaluation set:

  • Loss: 2.4481

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: 16
  • eval_batch_size: 2
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 2
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 128
  • total_eval_batch_size: 4
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_ratio: 0.03
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss
2.9976 0.9999 2548 2.8137
2.5794 1.9998 5096 2.4651
2.5149 2.9997 7644 2.4481

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.1.2+cu121
  • Datasets 2.14.6
  • Tokenizers 0.19.1
Downloads last month
20
Safetensors
Model size
162M params
Tensor type
BF16
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for kykim0/pythia-160m-tulu-v2-mix

Finetuned
(77)
this model

Dataset used to train kykim0/pythia-160m-tulu-v2-mix