Sign in to view source links and access this dataset
Description
ShahzebKhoso created this dataset on Hugging Face, last updated on 2026-05-29. It contains raw evaluation metrics, execution telemetry logs, and structural syntax outputs from running the Mostly Basic Python Problems (MBPP) benchmark against the DeepSeek R1 1.5B model. The data establishes performance boundaries for lightweight reasoning models under strict execution time limits on consumer hardware.
Use Cases
Benchmarking lightweight reasoning models based on MBPP performance metrics
Analyzing execution telemetry logs to understand model behavior under time constraints
Comparing structural syntax outputs for code generation quality assessment
Evaluating the performance of distilled reasoning architectures on consumer hardware
Strengths
Includes raw evaluation metrics, execution telemetry logs, and structural syntax outputs
Specifically benchmarks the DeepSeek R1 1.5B distilled reasoning architecture
Focuses on performance under strict execution time limits on consumer hardware
Limitations
Description metadata is limited; actual data quality requires manual inspection after download
Column-level documentation is absent; field semantics must be inferred after download
Row count is unknown, which may limit suitability assessment
Provenance
Source
Hugging Face user ShahzebKhoso
Collection Method
Running the MBPP benchmark against the DeepSeek R1 1.5B model
Freshness
Last updated 2026-05-29 06:14:10; freshness should be verified
License is unknown; terms of use must be verified before application.