Spaces:
Running
Running
Commit History
-
update submission
burtenshaw04b4890 -
Update submissions.json (#10)
Verified signature- Update submissions.json (c8b8c2ccc6ee2201e57604fde28941942a2172f9) Co-authored-by: Malan Lakshan Evans <malanevans@users.noreply.huggingface.co>4c35f7c
-
fix dataframe sorting again
burtenshawbfcd720 -
fix fn
burtenshaw12ddea5 -
use function
burtenshaw4262482 -
tidy up table
burtenshaw157452e -
improve submission error handling
burtenshaw28362c5 -
Merge branch 'main' into pr/9
burtenshaw77db0c8 -
Merge branch 'main' into pr/8
burtenshaw167a1ee -
Merge branch 'main' into pr/6
burtenshaw2c92e91 -
Merge branch 'main' into pr/5
burtenshawcbd7058 -
Merge branch 'main' into pr/3
burtenshaw20fe512 -
fix json
burtenshawf55fa21 -
Merge branch 'pr/2'
burtenshawfff02d1 -
fix submission
burtenshawc12e4af -
Update submissions.json
Verified signatureRan on a notebook with: ```python !pip install uv -qqq !uv init !uv add "triton<=3.2.0" "vllm<0.10.2" lighteval[vllm] -qqq !uv run lighteval vllm "model_name=andregustavo/SmolLM3-Custom-SFT,dtype=float16" "lighteval|gsm8k|0|0" --push-to-hub --results-org andregustavo ``` This should work on hf jobs: ```python hf jobs uv run \ # run a hf jobs job with uv --flavor a10g-large \ # select the machine size --with "triton<=3.2.0" --with "vllm<0.10.2" --with "lighteval[vllm]" \ # install lighteval with vllm dependencies -s HF_TOKEN \ # share the huggingface write token lighteval vllm "model_name=andregustavo/SmolLM3-Custom-SFT,dtype=float16" "lighteval|gsm8k|0|0" --push-to-hub --results-org <your-username> ```da80607 -
Update submissions.json (#1)
Verified signature- Update submissions.json (11c9f5fdc087b4bf298348e901510909b94dbf7e) Co-authored-by: Harshwardhan Fartale <emharsha1812@users.noreply.huggingface.co>095acac
-
Update submissions.json
Verified signaturehf jobs uv run --flavor a10g-large --with 'vllm<=0.10' --with "lighteval[vllm]<=0.10" -s HF_TOKEN lighteval vllm "model_name=quablab/SmolLM3-Custom-SFT" "lighteval|gsm8k|0|0" --push-to-hub --results-org quablab01200f1
-
Update submissions.json
Verified signature``` hf jobs uv run \ --flavor "a10g-large" \ --secrets HF_TOKEN \ --with "lighteval[vllm]" \ --with 'vllm<=0.10' \ lighteval vllm "model_name=jweston/SmolLM3-Custom-SFT" "lighteval|gsm8k|0|0" --push-to-hub --results-org "jweston" ```0998c07
-
Update submissions.json
Verified signaturewith T4 Colab GPU. Cant train a model without LORA. Model trained with LORA is not able to validate with lighteval vllm. So Using accelerate !lighteval accelerate "model_name=sanjay-saatyaki/smol-train,dtype=float16" "lighteval|gsm8k|0|0" --push-to-hub --results-org sanjay-saatyakied6e11d
-
Update submissions.json
Verified signatureCommand used: `lighteval accelerate "model_name=iamgroot42/smollm3-Custom-SFT" "lighteval|gsm8k|0|0" --push-to-hub --results-org iamgroot42`69c36b1
-
Update submissions.json
Verified signaturehf jobs uv run \ --flavor a10g-large \ --with "lighteval[vllm]" \ s HF_TOKEN \ lighteval vllm "model_name=msquaredd/SmolLM3-Custom-SFT-20250910143319" "lighteval|gsm8k|0|0" --push-to-hub --results-org msquaredd921af49
-
Update docs.md
Verified signature46d71ea
-
update url
burtenshaw4d25207 -
update subs
burtenshaw1675837 -
update docs
burtenshaw0e0e8f7 -
update app to use lighteval format
burtenshaw1c7c01e -
use lighteval for evaluation
burtenshawb5eec3d -
update presentation of app and docs
burtenshaw8b0cd75
-
requirements
burtenshaw059733c -
first commit
burtenshaw362d13b -
initial commit
Verified signatured107807