Views
No views yet
model_type: llava_qwen
with added BEV compressor/projector modules) fine-tuned for spatial reasoning on
multi-camera autonomous-driving data (nuScenes).LLaVA-NeXT/scripts/train/, entry scripts stage*.sh)checkpoint-XXXX trainer states were not uploaded). config.json:_name_or_path still records
the original local path of the run each model was initialized from.llava-onevision-google_siglip-so400m-patch14-384-Qwen_Qwen2-7B-Instruct-ov
is abbreviated to ov below. Runs marked † were initialized from a run that was already
deleted before archiving (not recoverable).| Run | Initialized from | Steps | Epochs |
|---|---|---|---|
ov_stage_0908 | lmms-lab/llava-onevision-qwen2-7b-ov (lineage root) | 11500 | 4 |
ov_1025_stage_1_ad | 1023_stage_1_l † | 5056 | 2 |
ov_1027_stage_1_m | stage_1_s_1023 † | 3437 | 1 |
ov_1027_stage_1_l | ov_1027_stage_1_m | 2909 | 1 |
ov_1028_stage_1_ad | ov_1027_stage_1_l | 625 | 1 |
ov_1028_stage_2_s2 | 1028_stage_2_s † | 703 | 1 |
ov_1028_stage_3_s | ov_1028_stage_2_s2 | 365 | 1 |
ov_1028_stage_6_s | ov_1028_stage_3_s | 494 | 1 |
ov_1028_stage_236_s | ov_1028_stage_6_s | 117 | 1 |
ov_1028_stage_2_l3 | ov_1028_stage_236_s | 656 | 1 |
ov_1028_stage_3_l2 | 1028_stage_2_l2 † | 500 | 1 |
ov_1029_stage_6_l2 | ov_1028_stage_3_l2 | 281 | 1 |
ov_1031_stage_6_ad | 1031_stage_6_b † | 351 | 1 |
ov_1031_stage_6_ad5 | 1031_stage_6_ad3 † | 1501 | 1 |