prog-y_tiny-llama-CDL-16

<!-- 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. -->

progressive-yoco-tiny-llama-CDL-16

This model is a fine-tuned version of /ephemeral/hossein/output/progressive-yoco-tiny-llama-CDL-17 on the reformatted_ultrachat_200k, the reformatted_MathInstruct and the small_slim_pajama datasets.

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: 2e-05
• train_batch_size: 42
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 6
• total_train_batch_size: 2016
• 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: 0.005
• training_steps: 150

Training results

Framework versions

• Transformers 4.45.2
• Pytorch 2.5.1+cu124
• Datasets 3.1.0
• Tokenizers 0.20.3