adaptations from pluto

llava/eval/model_vqa_ben.py

llava/model/multimodal_encoder/clip_encoder.py

scripts/eval/eval_vqa_llava.sh

scripts/eval/eval_vqa_ms.sh

scripts/eval/eval_vqa_rgb.sh

scripts/finetune/lora_ms_vqa_captions.sh

scripts/finetune/lora_rgb_vqa_captions.sh

scripts/finetune/ms_vqa_captions.sh

scripts/finetune/rgb_vqa_captions.sh

scripts/pretrain/pretrain_rgb.sh

slurm/check_gpu.py

slurm/lora_job.sh

slurm/ms_job.sh

slurm/rgb_job.sh