Skip to content

Fix eval max length #297

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 2 commits into
base: main
Choose a base branch
from
Open

Fix eval max length #297

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
e60d3f7
Update evaluate.py
Some-random Feb 12, 2025
f4909fe
Update README.md
Some-random Feb 12, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 6 additions & 6 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -193,7 +193,7 @@ We use `lighteval` to evaluate models, with custom tasks defined in `src/open_r1

```shell
MODEL=deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
MODEL_ARGS="pretrained=$MODEL,dtype=bfloat16,max_model_length=32768,gpu_memory_utilisation=0.8"
MODEL_ARGS="pretrained=$MODEL,dtype=bfloat16,max_model_length=4096,gpu_memory_utilisation=0.8"
OUTPUT_DIR=data/evals/$MODEL

# AIME 2024
Expand All @@ -219,14 +219,14 @@ lighteval vllm $MODEL_ARGS "custom|$TASK|0|0" \
```

> [!IMPORTANT]
> You must set `max_model_length=32768` in the `vllm` command to align with the `generation_size` we define per eval. Without this, `lighteval` will throw an error.
> You must set `max_model_length=4096` in the `vllm` command to align with the `generation_size` we define per eval. Without this, `lighteval` will throw an error.

To increase throughput across multiple GPUs, use _data parallel_ as follows:

```shell
NUM_GPUS=8
MODEL=deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
MODEL_ARGS="pretrained=$MODEL,dtype=bfloat16,data_parallel_size=$NUM_GPUS,max_model_length=32768,gpu_memory_utilisation=0.8"
MODEL_ARGS="pretrained=$MODEL,dtype=bfloat16,data_parallel_size=$NUM_GPUS,max_model_length=4096,gpu_memory_utilisation=0.8"
TASK=aime24
OUTPUT_DIR=data/evals/$MODEL

Expand All @@ -241,7 +241,7 @@ For large models which require sharding across GPUs, use _tensor parallel_ and r
```shell
NUM_GPUS=8
MODEL=deepseek-ai/DeepSeek-R1-Distill-Qwen-32B
MODEL_ARGS="pretrained=$MODEL,dtype=bfloat16,tensor_parallel_size=$NUM_GPUS,max_model_length=32768,gpu_memory_utilisation=0.8"
MODEL_ARGS="pretrained=$MODEL,dtype=bfloat16,tensor_parallel_size=$NUM_GPUS,max_model_length=4096,gpu_memory_utilisation=0.8"
TASK=aime24
OUTPUT_DIR=data/evals/$MODEL

Expand Down Expand Up @@ -295,7 +295,7 @@ To reproduce these results use the following command:
```shell
NUM_GPUS=1 # Set to 8 for 32B and 70B models
MODEL=deepseek-ai/{model_name}
MODEL_ARGS="pretrained=$MODEL,dtype=bfloat16,max_model_length=32768,gpu_memory_utilisation=0.8,tensor_parallel_size=$NUM_GPUS"
MODEL_ARGS="pretrained=$MODEL,dtype=bfloat16,max_model_length=4096,gpu_memory_utilisation=0.8,tensor_parallel_size=$NUM_GPUS"
OUTPUT_DIR=data/evals/$MODEL

lighteval vllm $MODEL_ARGS "custom|math_500|0|0" \
Expand Down Expand Up @@ -328,7 +328,7 @@ To reproduce these results use the following command:
```shell
NUM_GPUS=1 # Set to 8 for 32B and 70B models
MODEL=deepseek-ai/{model_name}
MODEL_ARGS="pretrained=$MODEL,dtype=bfloat16,max_model_length=32768,gpu_memory_utilisation=0.8,tensor_parallel_size=$NUM_GPUS"
MODEL_ARGS="pretrained=$MODEL,dtype=bfloat16,max_model_length=4096,gpu_memory_utilisation=0.8,tensor_parallel_size=$NUM_GPUS"
OUTPUT_DIR=data/evals/$MODEL

lighteval vllm $MODEL_ARGS "custom|gpqa:diamond|0|0" \
Expand Down
8 changes: 4 additions & 4 deletions src/open_r1/evaluate.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -102,7 +102,7 @@ def gpqa_prompt_fn(line, task_name: str = None):
evaluation_splits=["train"],
few_shots_split=None,
few_shots_select=None,
generation_size=32768,
generation_size=4096,
metric=[expr_gold_metric],
version=1,
)
Expand All @@ -117,7 +117,7 @@ def gpqa_prompt_fn(line, task_name: str = None):
evaluation_splits=["train"],
few_shots_split=None,
few_shots_select=None,
generation_size=32768,
generation_size=4096,
metric=[expr_gold_metric],
version=1,
)
Expand All @@ -131,7 +131,7 @@ def gpqa_prompt_fn(line, task_name: str = None):
evaluation_splits=["test"],
few_shots_split=None,
few_shots_select=None,
generation_size=32768,
generation_size=4096,
metric=[latex_gold_metric],
version=1,
)
Expand All @@ -145,7 +145,7 @@ def gpqa_prompt_fn(line, task_name: str = None):
evaluation_splits=["train"],
few_shots_split=None,
few_shots_select=None,
generation_size=32768, # needed for reasoning models like R1
generation_size=4096, # needed for reasoning models like R1
metric=[gpqa_metric],
stop_sequence=[], # no stop sequence, will use eos token
trust_dataset=True,
Expand Down