Public scan — anyone with this URL can view this analysis. Sign up to track your own repos privately, run scheduled re-scans, and get AI fix prompts via your dashboard.
Sign up free Scan another repo

yushengsu-thu/tune-lora-perf

https://github.com/yushengsu-thu/tune-lora-perf · scanned 2026-06-16 01:09 UTC (2 months, 1 week ago)

16 raw signals (0 security + 16 graph)

File as issue Image
UNIFIED Repobility · multi-layer engine · AI coders

Complete repo analysis

Last scanned 2 months, 1 week ago · v1 · 12 actionable findings from 1 signal source. 4 repeated signals grouped for readability. Security checks, system graph analysis, and verified AI-agent feedback are merged into one review queue.

JSON
Combined
86.6
/100
Security checks
87
0 findings
System graph
87
77.8% cov · 12 findings
Critical
0
click to filter
Agents
10
0 votes
Crowd
0
reported
Severity distribution — click a segment to filter
Active filters: excluding tests × Reset all
Severity: Critical 0 High 0 Medium 3 Low 6 9-Layer: Software Security Quality Integrity Frontend Hardware Data Network Cicd Source: Security checks 0 System graph 12 Crowd 0 Layer: Frontend 1 Security 1 Cicd 1 Quality 9

All 190 nodes from the latest scan, grouped by kind. Each node is a unit the engine identified (file, function, endpoint, table…). Most users won't need this view — it's primarily for debugging the engine's graph extraction or for AI agents that want to enumerate the project structure.

LabelLayerStatusPath
skill.md software healthy skill.md
README.md software healthy README.md
E2E_FULL_TEST_RUNBOOK.md software healthy E2E_FULL_TEST_RUNBOOK.md
pod.yaml software healthy pod.yaml
run_script.sh software healthy run_script.sh
skill_kernel_fusion.md software healthy skill_kernel_fusion.md
reformat_sanity.sh software healthy e2e_test_scripts/reformat_sanity.sh
gsm8k_lora.py software healthy e2e_test_scripts/gsm8k_lora.py
README.md software healthy e2e_test_scripts/README.md
fmla_import_chk.sh software healthy e2e_test_scripts/fmla_import_chk.sh
Qwen3.5-35B-A3B-FP8_sglnolora.sh software healthy e2e_test_scripts/Qwen3.5-35B-A3B-FP8_sglnolora.sh
Qwen3.5-35B-A3B-FP8_reformat_chk.sh software healthy e2e_test_scripts/Qwen3.5-35B-A3B-FP8_reformat_chk.sh
jit_chk_temp.sh software healthy e2e_test_scripts/jit_chk_temp.sh
bench_report.py software healthy e2e_test_scripts/bench_report.py
prompts_check.py software healthy e2e_test_scripts/prompts_check.py
flo-Qwen3.5-35B-A3B-FP8-tp1-pods.yaml software healthy e2e_test_scripts/gb200/flo-Qwen3.5-35B-A3B-FP8-tp1-pods.yaml
README.md software healthy e2e_test_scripts/gb200/README.md
Kimi-K2.5-NVFP4_run.sh software healthy e2e_test_scripts/gb200/Kimi-K2.5-NVFP4_run.sh
Kimi-K2.5-NVFP4-rf.yaml software healthy e2e_test_scripts/gb200/Kimi-K2.5-NVFP4-rf.yaml
Qwen3.5-35B-A3B-FP8_base.sh software healthy e2e_test_scripts/gb200/Qwen3.5-35B-A3B-FP8_base.sh
Qwen3.5-35B-A3B-FP8_run.sh software healthy e2e_test_scripts/gb200/Qwen3.5-35B-A3B-FP8_run.sh
Qwen3.5-35B-A3B-FP8_tp1_v2.sh software healthy e2e_test_scripts/gb200/Qwen3.5-35B-A3B-FP8_tp1_v2.sh
Kimi-K2.5-NVFP4_run_gb300.sh software healthy e2e_test_scripts/gb300/Kimi-K2.5-NVFP4_run_gb300.sh
rerun_bs128.sh software healthy e2e_test_scripts/gb300/rerun_bs128.sh
Qwen3-30B-A3B-Instruct-2507-BF16-expert_shared_run_gb300.sh software healthy e2e_test_scripts/gb300/Qwen3-30B-A3B-Instruct-2507-BF16-exp…
README.md software healthy e2e_test_scripts/gb300/README.md
Qwen3.5-35B-A3B-FP8_base_gb300.sh software healthy e2e_test_scripts/gb300/Qwen3.5-35B-A3B-FP8_base_gb300.sh
Kimi-K2.5-NVFP4-gb300.yaml software healthy e2e_test_scripts/gb300/Kimi-K2.5-NVFP4-gb300.yaml
jit_cache.sh software healthy e2e_test_scripts/gb300/jit_cache.sh
Qwen3.5-35B-A3B-FP8_run_gb300.sh software healthy e2e_test_scripts/gb300/Qwen3.5-35B-A3B-FP8_run_gb300.sh
Qwen3.5-35B-A3B-FP8-tp1-pods.yaml software healthy e2e_test_scripts/gb300/Qwen3.5-35B-A3B-FP8-tp1-pods.yaml
Qwen3.5-35B-A3B-FP8-pod.yaml software healthy e2e_test_scripts/gb300/Qwen3.5-35B-A3B-FP8-pod.yaml
Qwen3.5-35B-A3B-FP8_tp1_gb300.sh software healthy e2e_test_scripts/gb300/Qwen3.5-35B-A3B-FP8_tp1_gb300.sh
Kimi-K2.5-NVFP4_tests_only.sh software healthy e2e_test_scripts/gb300/Kimi-K2.5-NVFP4_tests_only.sh
RESULTS.md software healthy e2e_test_scripts/gb300/results/RESULTS.md
gpu_busy_witness.py software healthy dev/gpu_busy_witness.py
6_upload_results.sh software healthy dev/6_upload_results.sh
7_broadcast_jit_cache.sh software healthy dev/7_broadcast_jit_cache.sh
run_all.sh software healthy dev/run_all.sh
test_bf16_fold_p3.py software healthy dev/test_bf16_fold_p3.py
README.md software healthy dev/README.md
8_save_jit_cache.sh software healthy dev/8_save_jit_cache.sh
bench_profile_matrix.sh software healthy dev/bench_profile_matrix.sh
jit_store.sh software healthy dev/jit_store.sh
test_bf16_fold_p1.py software healthy dev/test_bf16_fold_p1.py
test_bf16_fold_ref.py software healthy dev/test_bf16_fold_ref.py
check_fused_align_equiv.py software healthy dev/check_fused_align_equiv.py
gen_dummy_lora.py software healthy dev/gen_dummy_lora.py
convert_to_perfetto_compatible.py software healthy dev/convert_to_perfetto_compatible.py
3_run_benchmark.sh software healthy dev/3_run_benchmark.sh

Showing first 50 of this kind. Full payload available via the JSON button at the top of the page.

LabelLayerStatusPath
main software healthy e2e_test_scripts/gsm8k_lora.py:7
one software healthy e2e_test_scripts/gsm8k_lora.py:15
post software healthy e2e_test_scripts/prompts_check.py:25
main software healthy e2e_test_scripts/prompts_check.py:35
ct software healthy e2e_test_scripts/prompts_check.py:57
chat software healthy e2e_test_scripts/prompts_check.py:60
comp software healthy e2e_test_scripts/prompts_check.py:64
gen software healthy e2e_test_scripts/prompts_check.py:68
c software healthy e2e_test_scripts/prompts_check.py:76
bench software healthy dev/test_bf16_fold_p3.py:46
build_case software healthy dev/test_bf16_fold_p1.py:20
build_expert_token_map software healthy dev/check_fused_align_equiv.py:33
run_case software healthy dev/check_fused_align_equiv.py:56
main software healthy dev/check_fused_align_equiv.py:133
normalize_base_name software healthy dev/gen_dummy_lora.py:74
log software healthy dev/gen_dummy_lora.py:94
already_done software healthy dev/gen_dummy_lora.py:98
module_suffix software healthy dev/gen_dummy_lora.py:104
base_weight_shapes software healthy dev/gen_dummy_lora.py:110
main software healthy dev/gen_dummy_lora.py:146
bucket software healthy dev/gen_dummy_lora.py:162
init_A software healthy dev/gen_dummy_lora.py:178
init_B software healthy dev/gen_dummy_lora.py:183
main software healthy dev/convert_to_perfetto_compatible.py:12
_process_events software healthy dev/convert_to_perfetto_compatible.py:30
_is_interest_event software healthy dev/convert_to_perfetto_compatible.py:44
build_routing software healthy dev/test_bf16_fold_p2.py:23
load software healthy dev/sanity_check_opt.py:34
phase_windows software healthy dev/sanity_check_opt.py:39
main software healthy dev/sanity_check_opt.py:47
fail software healthy regression/scripts/build_readme.py:16
_cell_sort_key software healthy regression/scripts/build_readme.py:53
tput software healthy regression/scripts/build_readme.py:62
sanity_status software healthy regression/scripts/build_readme.py:75
acc_vs_files software healthy regression/scripts/build_readme.py:95
prompts_status software healthy regression/scripts/build_readme.py:108
trace_counts software healthy regression/scripts/build_readme.py:118
has_subdir software healthy regression/scripts/build_readme.py:128
env software healthy regression/scripts/summary.py:13
load_lp software healthy regression/scripts/summary.py:39
last_row software healthy regression/scripts/summary.py:43
pctl software healthy regression/scripts/summary.py:48
split software healthy regression/scripts/summary.py:81
_server_decode_median software healthy regression/scripts/summary.py:101
post software healthy regression/scripts/prompts_check.py:25
main software healthy regression/scripts/prompts_check.py:35
ct software healthy regression/scripts/prompts_check.py:57
chat software healthy regression/scripts/prompts_check.py:60
comp software healthy regression/scripts/prompts_check.py:64
gen software healthy regression/scripts/prompts_check.py:68

Showing first 50 of this kind. Full payload available via the JSON button at the top of the page.

LabelLayerStatusPath
e2e_test_scripts software healthy e2e_test_scripts
gb200 software healthy e2e_test_scripts/gb200
gb300 software healthy e2e_test_scripts/gb300
results software healthy e2e_test_scripts/gb300/results
dev software healthy dev
models software healthy dev/models
Qwen3.5-35B-A3B-FP8-expert_shared software healthy dev/models/Qwen3.5-35B-A3B-FP8-expert_shared
Qwen3-30B-A3B-Instruct-2507-BF16-expert_shared software healthy dev/models/Qwen3-30B-A3B-Instruct-2507-BF16-expert_shared
Qwen3-30B-A3B-Instruct-2507-BF16 software healthy dev/models/Qwen3-30B-A3B-Instruct-2507-BF16
regression software healthy regression
gb200 software healthy regression/gb200
models software healthy regression/gb200/models
Qwen3.5-35B-A3B-FP8 software healthy regression/gb200/models/Qwen3.5-35B-A3B-FP8
Kimi-K2.5-NVFP4 software healthy regression/gb200/models/Kimi-K2.5-NVFP4
scripts software healthy regression/scripts
gb300 software healthy regression/gb300
models software healthy regression/gb300/models
Qwen3.5-35B-A3B-FP8 software healthy regression/gb300/models/Qwen3.5-35B-A3B-FP8
Kimi-K2.5-NVFP4 software healthy regression/gb300/models/Kimi-K2.5-NVFP4

LabelLayerStatusPath
k8s::Pod::pod.yaml hardware healthy pod.yaml
k8s::Pod::e2e_test_scripts/gb200/flo-Qwen3.5-35B-A3B-FP8-tp… hardware healthy e2e_test_scripts/gb200/flo-Qwen3.5-35B-A3B-FP8-tp1-pods.yaml
k8s::Pod::e2e_test_scripts/gb200/Kimi-K2.5-NVFP4-rf.yaml hardware healthy e2e_test_scripts/gb200/Kimi-K2.5-NVFP4-rf.yaml
k8s::Pod::e2e_test_scripts/gb300/Kimi-K2.5-NVFP4-gb300.yaml hardware healthy e2e_test_scripts/gb300/Kimi-K2.5-NVFP4-gb300.yaml
k8s::Pod::e2e_test_scripts/gb300/Qwen3.5-35B-A3B-FP8-tp1-po… hardware healthy e2e_test_scripts/gb300/Qwen3.5-35B-A3B-FP8-tp1-pods.yaml
k8s::Pod::e2e_test_scripts/gb300/Qwen3.5-35B-A3B-FP8-pod.ya… hardware healthy e2e_test_scripts/gb300/Qwen3.5-35B-A3B-FP8-pod.yaml
k8s::Pod::dev/models/Qwen3-30B-A3B-Instruct-2507-BF16-exper… hardware healthy dev/models/Qwen3-30B-A3B-Instruct-2507-BF16-expert_shared/p…
k8s::Pod::dev/models/Qwen3-30B-A3B-Instruct-2507-BF16/pod.y… hardware healthy dev/models/Qwen3-30B-A3B-Instruct-2507-BF16/pod.yaml
k8s::Pod::regression/gb200/models/Qwen3.5-35B-A3B-FP8/pod.y… hardware healthy regression/gb200/models/Qwen3.5-35B-A3B-FP8/pod.yaml
k8s::Pod::regression/gb200/models/Kimi-K2.5-NVFP4/pod.yaml hardware healthy regression/gb200/models/Kimi-K2.5-NVFP4/pod.yaml
k8s::Pod::regression/gb300/models/Qwen3.5-35B-A3B-FP8/pod.y… hardware healthy regression/gb300/models/Qwen3.5-35B-A3B-FP8/pod.yaml
k8s::Pod::regression/gb300/models/Kimi-K2.5-NVFP4/pod.yaml hardware healthy regression/gb300/models/Kimi-K2.5-NVFP4/pod.yaml

LabelLayerStatusPath
port:30000 network healthy e2e_test_scripts/Qwen3.5-35B-A3B-FP8_sglnolora.sh
port:20000 network healthy e2e_test_scripts/gb200/Kimi-K2.5-NVFP4-rf.yaml
port:30001 network healthy e2e_test_scripts/gb200/Qwen3.5-35B-A3B-FP8_tp1_v2.sh
port:256 network healthy regression/gb200/models/Qwen3.5-35B-A3B-FP8/pod.yaml
port:97 network healthy regression/gb200/models/Qwen3.5-35B-A3B-FP8/pod.yaml

LabelLayerStatusPath
k8s::ComputeDomain::e2e_test_scripts/gb200/Kimi-K2.5-NVFP4-… hardware healthy e2e_test_scripts/gb200/Kimi-K2.5-NVFP4-rf.yaml
k8s::ComputeDomain::e2e_test_scripts/gb300/Kimi-K2.5-NVFP4-… hardware healthy e2e_test_scripts/gb300/Kimi-K2.5-NVFP4-gb300.yaml
k8s::ComputeDomain::regression/gb200/models/Kimi-K2.5-NVFP4… hardware healthy regression/gb200/models/Kimi-K2.5-NVFP4/pod.yaml
k8s::ComputeDomain::regression/gb300/models/Kimi-K2.5-NVFP4… hardware healthy regression/gb300/models/Kimi-K2.5-NVFP4/pod.yaml

LabelLayerStatusPath
repobility-clone-rtro8qkf software healthy /tmp/repobility-clone-rtro8qkf

LabelLayerStatusPath
127.0.0.1 network healthy run_script.sh

LabelLayerStatusPath
vps::gcp hardware healthy README.md

LabelLayerStatusPath
gpu (detected) hardware healthy pod.yaml
For AI agents: Voting guide (TP/FP) MCP manifest Stdio wrapper SARIF Integrate Findings queue Vote TP/FP on findings to calibrate the engine.
For AI agents + API integrations
Email me when this repo regresses
Free. We re-scan periodically; new criticals → your inbox. No signup required for the scan itself.
API access

This page is publicly accessible at: https://repobility.com/scan/efec7f71-1742-409f-8db9-9410b611d7f3/

To check status programmatically (no auth required): 

curl -s https://repobility.com/api/v1/public/scan/efec7f71-1742-409f-8db9-9410b611d7f3/

Important — please don't re-submit the same URL repeatedly. The submission endpoint is idempotent: re-submitting the same git URL returns this same scan_token, not a new one. To re-scan this repo, sign up free and use the dashboard.