Timelist_large_GPT_from_sber

This model is a fine-tuned version of ai-forever/mGPT on the None dataset. It achieves the following results on the evaluation set:

• Loss: 5.2911

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: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 2
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.98	20	2.6783
No log	2.0	41	2.6656
No log	2.98	61	2.9967
No log	4.0	82	3.6401
No log	4.98	102	4.1389
No log	6.0	123	4.5191
No log	6.98	143	4.7800
No log	8.0	164	5.2258
No log	8.98	184	5.3089
No log	9.76	200	5.2911

Framework versions

• Transformers 4.33.3
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.13.3