omnisvg-n1-image-finetune-nopenalty-097b037-32icons-9b9b1699
completed —
Measured result
Verbatim from the run's summary.json.
checks
clip_gain_over_control
false
ended_rate
false
reference_top1_rate
false
clip
repo
openai/clip-vit-base-patch32
revision
3d74acf9a28c67741b2f4f2ea7635f0aaf6f0268
clip_to_caption
mean
0.23648546659387648
median
0.2310640886425972
n
64
clip_to_reference
mean
0.8642141604796052
median
0.8629124760627747
n
64
codec_valid_rate
0.125
config_sha256
d2bd0a77284ed948…control
clip_to_reference_mean
0.8642141604796052
control_clip_to_reference_mean
0.9046997493132949
control_codec_valid_rate
0.140625
control_reference_top1_rate
0.609375
control_rows_sha256
5cabd027e8c4e617…gain_ci95
-0.06706393379718065, -0.014240840589627626
icons_improved
8
mean_gain
-0.04048558883368969
paired_icons
32
report_root
reports/learning/omnisvg-n1-image-control
criteria
clip_gain_over_control_ci_excludes_zero
true
min_ended_rate
0.9
min_reference_top1_rate
0.75
decoded_rate
1.0
drawings
64
ended_rate
0.109375
failures
encode:program needs 1 contours and up to 1022 segments
1
encode:program needs 1 contours and up to 251 segments
1
encode:program needs 1 contours and up to 506 segments
1
encode:program needs 1 contours and up to 511 segments
7
encode:program needs 1 contours and up to 512 segments
1
encode:program needs 1 contours and up to 514 segments
1
encode:program needs 1 contours and up to 537 segments
1
encode:program needs 1 contours and up to 834 segments
1
encode:program needs 12 contours and up to 462 segments
1
encode:program needs 127 contours and up to 7 segments
1
encode:program needs 2 contours and up to 1014 segments
1
encode:program needs 2 contours and up to 501 segments
2
encode:program needs 2 contours and up to 506 segments
1
encode:program needs 2 contours and up to 978 segments
1
encode:program needs 20 contours and up to 455 segments
1
encode:program needs 3 contours and up to 461 segments
1
encode:program needs 3 contours and up to 483 segments
1
encode:program needs 3 contours and up to 494 segments
1
encode:program needs 3 contours and up to 497 segments
1
encode:program needs 35 contours and up to 19 segments
1
encode:program needs 4 contours and up to 487 segments
1
encode:program needs 44 contours and up to 14 segments
1
encode:program needs 47 contours and up to 16 segments
1
encode:program needs 5 contours and up to 628 segments
1
encode:program needs 52 contours and up to 11 segments
2
encode:program needs 6 contours and up to 437 segments
1
encode:program needs 6 contours and up to 445 segments
1
encode:program needs 6 contours and up to 451 segments
1
encode:program needs 6 contours and up to 462 segments
1
encode:program needs 62 contours and up to 15 segments
1
encode:program needs 63 contours and up to 11 segments
1
encode:program needs 64 contours and up to 10 segments
1
encode:program needs 73 contours and up to 17 segments
1
encode:program needs 76 contours and up to 11 segments
1
encode:program needs 79 contours and up to 8 segments
2
encode:program needs 83 contours and up to 12 segments
1
encode:program needs 85 contours and up to 13 segments
1
encode:program needs 86 contours and up to 7 segments
1
encode:program needs 9 contours and up to 435 segments
1
normalize:geometry_error: M has sets of 2 args, 1 invalid
2
pack:program needs 148 packed segments but capacity is 128
1
pack:program needs 179 packed segments but capacity is 128
1
pack:program needs 186 packed segments but capacity is 128
1
pack:program needs 209 packed segments but capacity is 128
1
pack:program needs 236 packed segments but capacity is 128
1
pack:program needs 573 packed segments but capacity is 128
1
icons
32
max_new_tokens
2048
mean_snap_distance_rgb
4.572507278980439
median_paths
6.5
median_tokens
2048.0
model
adapter
data/processed/prior/omnisvg-n1-image-finetune/adapter.zip
adapter_sha256
15b8be9922c3a678…checkpoint_sha256
39f36759fc677b32…fine_tuned
true
parameters
3846914048
repo
OmniSVG/OmniSVG1.1_4B
revision
117d4c4541839e02943554fe71c0a0d54fcef819
predeclared_outcome
falsified
prompt_style
image
reference_rank
chance_top1
0.03125
mean
7.140625
scored
64
top1_rate
0.359375
top5_rate
0.625
rows_sha256
818c792a817c82b9…samples_per_icon
2
sampling
repetition_penalty
1.0
temperature
0.3
top_k
50
top_p
0.9
schema_version
1
sheet_sha256
af1433925c877842…study_version
omnisvg-n1-image-finetune-nopenalty
timing
generate_seconds
2386.5209589300503
load_seconds
37.46009345600032
peak_vram_gib
8.081450462341309
seconds_per_drawing
37.289389983282035
Visual output
State transitions
- planned2026-09-22T16:19:56Z
- completed2026-09-22T18:48:34Zwithout the penalty the adapter draws even longer: 11% end, median 2,048 tokens, the right icon first for 23 of 64 (0.36), similarity 0.864, paired gain -0.041 with an interval of -0.067 to -0.014
Run record
This run has no run.yaml. What follows is the identity and configuration carried by its rows in state/runs.jsonl, the append-only registry.
run_id
omnisvg-n1-image-finetune-nopenalty-097b037-32icons-9b9b1699
config
configs/learning/omnisvg-n1-image-finetune-nopenalty.yaml
outputs
report_root
reports/learning/omnisvg-n1-image-finetune-nopenalty