gemma2b-summarize-claude3sonnet-8k

This model is a fine-tuned version of google/gemma-2b on the llama-duo/synth_summarize_dataset_dedup dataset. It achieves the following results on the evaluation set:

• Loss: 2.5421

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: 0.0002
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 3
• gradient_accumulation_steps: 2
• total_train_batch_size: 48
• total_eval_batch_size: 24
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss
1.9711	0.9804	25	2.6222
1.3473	2.0	51	2.5571
1.1845	2.9804	76	2.5050
1.1242	4.0	102	2.5089
1.0937	4.9804	127	2.5239
1.0632	6.0	153	2.5275
1.0377	6.9804	178	2.5316
1.0278	8.0	204	2.5399
1.0176	8.9804	229	2.5422
1.007	9.8039	250	2.5421

Framework versions

• PEFT 0.10.0
• Transformers 4.40.0
• Pytorch 2.2.2+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1