hallucination_bin_detector_v4.3

This model is a fine-tuned version of IlyaGusev/saiga_yandexgpt_8b on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2487

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: 1
• eval_batch_size: 1
• seed: 1337
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 8
• total_train_batch_size: 64
• total_eval_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.05
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
0.3179	0.1761	16	0.3490
0.2609	0.3521	32	0.3359
0.261	0.5282	48	0.2961
0.2741	0.7043	64	0.2702
0.2638	0.8803	80	0.2857
0.1861	1.0564	96	0.2860
0.2338	1.2325	112	0.2499
0.1728	1.4085	128	0.2424
0.2371	1.5846	144	0.2385
0.2579	1.7607	160	0.2567
0.1639	1.9367	176	0.2410
0.1608	2.1128	192	0.2580
0.1984	2.2889	208	0.2476
0.1507	2.4649	224	0.2434
0.2221	2.6410	240	0.2481
0.1298	2.8171	256	0.2487

Framework versions

• PEFT 0.9.0
• Transformers 4.45.2
• Pytorch 2.3.0a0+6ddf5cf85e.nv24.04
• Datasets 4.4.1
• Tokenizers 0.20.3