gut_1024-finetuned-lora-bert-base-lastln-t2t

This model is a fine-tuned version of AIRI-Institute/gena-lm-bert-base-lastln-t2t on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4049
  • F1: 0.8657
  • Mcc Score: 0.6421
  • Accuracy: 0.8290

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: 0.0005
  • train_batch_size: 8
  • eval_batch_size: 64
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • training_steps: 1000

Training results

Training Loss Epoch Step Validation Loss F1 Mcc Score Accuracy
0.6139 0.02 100 0.5419 0.7847 0.4843 0.7487
0.5575 0.04 200 0.5407 0.7875 0.4941 0.7530
0.5339 0.05 300 0.5027 0.8347 0.5669 0.7943
0.4807 0.07 400 0.4584 0.8551 0.6032 0.8053
0.5071 0.09 500 0.4408 0.8583 0.6192 0.8180
0.4011 0.11 600 0.4346 0.8576 0.6127 0.8129
0.4133 0.12 700 0.4694 0.8461 0.6007 0.8100
0.424 0.14 800 0.4073 0.8651 0.6357 0.8239
0.4084 0.16 900 0.3974 0.8677 0.6429 0.8264
0.4427 0.18 1000 0.4049 0.8657 0.6421 0.8290

Framework versions

  • Transformers 4.38.1
  • Pytorch 2.1.0+cu121
  • Datasets 2.17.1
  • Tokenizers 0.15.2
Downloads last month
22
Safetensors
Model size
139M params
Tensor type
I64
·
F32
·
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 LiukG/gut_1024-finetuned-lora-bert-base-lastln-t2t

Finetuned
(2)
this model