hrezaei/flan-t5la-large
023
<!-- 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. -->
flan-t5la-large
This model is a fine-tuned version of hrezaei/flan-t5la-large on the HuggingFaceFW/fineweb sample-350BT dataset. It achieves the following results on the evaluation set:
- Perplexity: 5.0594
- Loss: 1.6212
- Accuracy: 0.0025
- Lookahead Perplexity: 22.5300
- Lookahead Loss: 3.1148
- Base Perplexity: 1.1402
- Base Loss: 0.1312
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
- trainbatchsize: 16
- evalbatchsize: 16
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- totaltrainbatch_size: 32
- totalevalbatch_size: 32
- optimizer: Use adamwtorchfused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lrschedulertype: linear
- training_steps: 524288
Training results
Framework versions
- Transformers 4.57.0.dev0
- Pytorch 2.8.0+cu128
- Datasets 4.2.0
- Tokenizers 0.22.1
