prev_work_15B_b256_80gb_H100_seed_41

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

• Loss: 2.9952

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: 0.0035355339059327372
• train_batch_size: 256
• eval_batch_size: 256
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 2048
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 1000
• num_epochs: 1
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
3.8481	0.0861	5000	3.5245
3.4627	0.1722	10000	3.4316
3.3956	0.2582	15000	3.3806
3.3522	0.3443	20000	3.3436
3.3146	0.4304	25000	3.3071
3.2775	0.5165	30000	3.2681
3.2372	0.6026	35000	3.2279
3.1951	0.6887	40000	3.1836
3.1493	0.7747	45000	3.1376
3.1002	0.8608	50000	3.0878
3.0481	0.9469	55000	3.0349

Framework versions

• Transformers 4.45.0.dev0
• Pytorch 2.6.0+cu124
• Datasets 3.2.0
• Tokenizers 0.19.1