results

This model is a fine-tuned version of facebook/bart-base on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.7266

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-05
  • train_batch_size: 4
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss
0.8268 0.08 1000 0.7354
0.7977 0.16 2000 0.7251
0.7739 0.24 3000 0.7259
0.771 0.32 4000 0.7269
0.7468 0.4 5000 0.7269
0.751 0.48 6000 0.7501
0.7483 0.56 7000 0.7502
0.7443 0.64 8000 0.7253
0.7294 0.72 9000 0.7309
0.7309 0.8 10000 0.7260
0.7424 0.88 11000 0.7304
0.7348 0.96 12000 0.7276
0.7421 1.04 13000 0.7327
0.7333 1.12 14000 0.7417
0.7444 1.2 15000 0.7296
0.7463 1.28 16000 0.7257
0.7324 1.3600 17000 0.7253
0.7297 1.44 18000 0.7314
0.7358 1.52 19000 0.7253
0.7442 1.6 20000 0.7248
0.7384 1.6800 21000 0.7388
0.7345 1.76 22000 0.7259
0.7218 1.8400 23000 0.7284
0.7426 1.92 24000 0.7253
0.7375 2.0 25000 0.7389
0.7443 2.08 26000 0.7305
0.7286 2.16 27000 0.7258
0.7269 2.24 28000 0.7264
0.7391 2.32 29000 0.7270
0.7377 2.4 30000 0.7283
0.7319 2.48 31000 0.7329
0.7352 2.56 32000 0.7254
0.7141 2.64 33000 0.7285
0.7317 2.7200 34000 0.7253
0.7334 2.8 35000 0.7305
0.7332 2.88 36000 0.7282
0.7309 2.96 37000 0.7266

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.0
  • Datasets 3.0.0
  • Tokenizers 0.19.1
Downloads last month
18
Safetensors
Model size
140M params
Tensor type
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 Pra-tham/results

Base model

facebook/bart-base
Finetuned
(368)
this model