README.md

metadata

library_name: transformers
license: apache-2.0
base_model: ai21labs/Jamba-tiny-dev
tags:
  - alignment-handbook
  - trl
  - sft
  - generated_from_trainer
  - trl
  - sft
  - generated_from_trainer
datasets:
  - bitext/Bitext-combined-banking-wealth_management-mortgage_loans
model-index:
  - name: BT-Jamba-tiny-dev-customer-support
    results: []

BT-Jamba-tiny-dev-customer-support

This model is a fine-tuned version of ai21labs/Jamba-tiny-dev on the bitext/Bitext-combined-banking-wealth_management-mortgage_loans dataset. It achieves the following results on the evaluation set:

• Loss: 0.5598

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: 4
• eval_batch_size: 4
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• total_eval_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.01
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
0.6095	1.0	150	0.6130
0.5353	2.0	300	0.5674
0.4782	3.0	450	0.5598

Framework versions

• Transformers 4.45.0.dev0
• Pytorch 2.1.0
• Datasets 2.18.0
• Tokenizers 0.19.1