llama2-7b-chat-hf-english-to-hinglish

This model is a fine-tuned version of NousResearch/Llama-2-7b-chat-hf on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.1746
• Rouge Scores: {'rouge1': 0.8885908254946812, 'rouge2': 0.7710863010584095, 'rougeL': 0.5959196285242199, 'rougeLsum': 0.8887611638083375}
• Bleu Scores: [0.9146534717527709, 0.8952839315431885, 0.8704729625819737, 0.8440396772968578]
• Gen Len: 2048.0

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
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge Scores	Bleu Scores	Gen Len
2.4347	1.0	19	2.2502	{'rouge1': 0.8978799395571857, 'rouge2': 0.7786106895702147, 'rougeL': 0.5929935988553734, 'rougeLsum': 0.8979590919999867}	[0.8924998190589056, 0.8765058761668494, 0.8529072060046804, 0.8271335845757618]	2048.0
2.1641	2.0	38	2.1817	{'rouge1': 0.8908622699691104, 'rouge2': 0.787877924739433, 'rougeL': 0.5912826901692394, 'rougeLsum': 0.8911134776621241}	[0.9026796342264707, 0.887688173576005, 0.8645415966459645, 0.839085075553431]	2048.0
2.085	3.0	57	2.1746	{'rouge1': 0.8885908254946812, 'rouge2': 0.7710863010584095, 'rougeL': 0.5959196285242199, 'rougeLsum': 0.8887611638083375}	[0.9146534717527709, 0.8952839315431885, 0.8704729625819737, 0.8440396772968578]	2048.0

Framework versions

• PEFT 0.9.1.dev0
• Transformers 4.37.0
• Pytorch 2.1.0+cu121
• Datasets 2.18.1.dev0
• Tokenizers 0.15.2