This model is a finetuned version of Llama2-7b-hf using a Dataset Feedback Collection with questions, answers, and evaluations related to general subjects.

It used a QLORA version for the finetuning process.

This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.

Use destinated to improve automatic healthcare evaluations. LLM as a Judge.

The dataset used is not specialized in healthcare questions, answers, and evaluation with rubric scores. The main risk is getting evaluations not so customized to our proposal.

Users (both direct and downstream) should be made aware of the risks, biases, and limitations of the model. More healthcare data is needed for further recommendations.

The prometheus-eval/Feedback-Collection dataset contains 9,996 rows. (90%) of those rows was randomized and used for training.

It wasn't a full finetuning, but only of the following linear layers of the transformer model: model: Percentage of trained parameters Trainable: 20971520 Total: 7262703616 Percentage: 0.2888% Linear Layers: ['k_proj', 'gate_proj', 'o_proj', 'q_proj', 'v_proj', 'down_proj', 'up_proj'].

For LORA step it was used the following hyperparameters: r=8, lora_alpha=32, target_modules=modules, lora_dropout=0.05

The hiperparameters related to the training process:

    per_device_train_batch_size=1,
    gradient_accumulation_steps=4,
    warmup_steps=0,
    max_steps=100,
    learning_rate=2e-4,
    logging_steps=20,
    output_dir="outputs",
    optim="paged_adamw_8bit",
    save_strategy="epoch"
    

It was used for the assessment of the training process Loss Function (built-in Cross-entropy for Transformers). It's the cross-entropy between the predicted probability matrix between the predicted probability matrix ∥sentence length∥×∥vocab∥ (right before taking the argmax to find the token to output), and the ∥sentence length∥ -length vector of token IDs as the true label. It is a method of evaluating how well your algorithm models your dataset. In this case, it's a method to assess how well a transformer predicts the evaluations for each pair question and generated answer. There is a decrease in the Loss function indicating a learning process. Ideally, adding more data and adjusting hyperparameters can close this function to zero.

The metrics related to the Finetuning process:

global_step=100 training_loss=0.5179 train_runtime: 632 s https://wandb.ai/eeg-neuko/huggingface/runs/ckdqyizl?nw=nwuserguicondor1512

It used a GPU A100 with VRAM of 40 GB from Google Colab Pro+.