BitChop: A Heuristic Approach to Memory Footprint Reduction in AI Training
Bibliographic record
Abstract
AI training costs have been greatly increasing in recent years, with bigger neural networks that require increased computational capabilities and memory storage. AI training is greatly limited by memory access time. We introduce BitChop, a heuristic-based lossy compression method that reduces the amount of memory that is being loaded and stored during training. BitChop dynamically adjusts the precision and format of the floating-point containers of the network's activations. BitChop adjusts the bit-lengths of mantissas based on different heuristics that observe changes in the loss function of the network. Across the tested heuristics, BitChop reduces the total mantissa footprint to 25% of baseline when applied over FP32, and to 19% when applied over BFloat16. Moreover, BitChop reduces the total footprint of the networks to an average of 46% over FP32 and to an average of 40% over BFloat16. These reductions in footprint result in 2x performance and 3.5x energy efficiency improvements.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".