107 kB

Ctrl+K

1 contributor

History: 6 commits

ProCreations

v2: Beam reasoning-effort levels (low/medium/high), Accurate variant; v1 is tagged

0052e99 verified 38 minutes ago

audit_eval.py

3.32 kB
v2: Beam reasoning-effort levels (low/medium/high), Accurate variant; v1 is tagged 38 minutes ago
build_dataset.py

5.75 kB
Dory-Beam-4B weights, config, tokenizer, license and code 3 days ago
build_effort_dataset.py

6.04 kB
v2: Beam reasoning-effort levels (low/medium/high), Accurate variant; v1 is tagged 38 minutes ago
build_evalsets.py

3.19 kB
Dory-Beam-4B weights, config, tokenizer, license and code 3 days ago
build_pool.py

5.08 kB
Dory-Beam-4B weights, config, tokenizer, license and code 3 days ago
build_space_data.py

3.23 kB
v2: Beam reasoning-effort levels (low/medium/high), Accurate variant; v1 is tagged 38 minutes ago
check_dory.py

8.05 kB
Dory-Beam-4B weights, config, tokenizer, license and code 3 days ago
export_effort_data.py

3 kB
v2: Beam reasoning-effort levels (low/medium/high), Accurate variant; v1 is tagged 38 minutes ago
export_eval_data.py

5.75 kB
v2: Beam reasoning-effort levels (low/medium/high), Accurate variant; v1 is tagged 38 minutes ago
export_release_data.py

3.26 kB
Real-time M4 Max race video + raw timings; Mac speed note; race and eval scripts 3 days ago
generate.py

8.74 kB
Dory-Beam-4B weights, config, tokenizer, license and code 3 days ago
make_chart.py

6.98 kB
v2: Beam reasoning-effort levels (low/medium/high), Accurate variant; v1 is tagged 38 minutes ago
make_chart_v1.py

4.69 kB
v2: Beam reasoning-effort levels (low/medium/high), Accurate variant; v1 is tagged 38 minutes ago
merge_continued.py

4.84 kB
Re-evaluate at a 64k-token limit: resume answers cut off at 32k (corrects AIME comparisons) about 10 hours ago
merge_dory.py

3.01 kB
Dory-Beam-4B weights, config, tokenizer, license and code 3 days ago
race_record.py

4.58 kB
v2: Beam reasoning-effort levels (low/medium/high), Accurate variant; v1 is tagged 38 minutes ago
race_render.py

7.88 kB
v2: Beam reasoning-effort levels (low/medium/high), Accurate variant; v1 is tagged 38 minutes ago
run_main.sh

739 Bytes
Dory-Beam-4B weights, config, tokenizer, license and code 3 days ago
spawn_continue.py

1.5 kB
Re-evaluate at a 64k-token limit: resume answers cut off at 32k (corrects AIME comparisons) about 10 hours ago
spawn_effort_evals.py

1.32 kB
v2: Beam reasoning-effort levels (low/medium/high), Accurate variant; v1 is tagged 38 minutes ago
summarize_effort.py

3.25 kB
v2: Beam reasoning-effort levels (low/medium/high), Accurate variant; v1 is tagged 38 minutes ago
train_dory.py

11.5 kB
Dory-Beam-4B weights, config, tokenizer, license and code 3 days ago
verify.py

1.73 kB
Dory-Beam-4B weights, config, tokenizer, license and code 3 days ago