Hugging Face's logo

DorianAtSchool
/
openvla-7b-vlabench-100k-steps-lora-finetuned

Safetensors
openvla
custom_code
Model card Files
xet
Community

openvla-7b base finetuned on VLABenchmark (rlds) using LoRA for 100,000k steps on 2 gpus.

https://wandb.ai/dbenhamougol-umass-amherst/openvla_vlabench_finetune/reports/openvla-7b-LoRA-finetune--VmlldzoxNDY3MzExNg?accessToken=65r72y9vn87onhwo4xoycj5vjao1lq4dei83bwegpz4x45itpc0nyce9pes5342k

Downloads last month
-
Safetensors
Model size
8B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for DorianAtSchool/openvla-7b-vlabench-100k-steps-lora-finetuned

Base model

openvla/openvla-7b
Finetuned
(16)
this model

Dataset used to train DorianAtSchool/openvla-7b-vlabench-100k-steps-lora-finetuned