Rethinking Optimization and Architecture for Tiny Language Models
Paper
•
2402.02791
•
Published
•
12
Model Training
-
-
More Agents Is All You Need
Paper • 2402.05120 • Published • 51 -
Scaling Laws for Forgetting When Fine-Tuning Large Language Models
Paper • 2401.05605 • Published -
Aligning Large Language Models with Counterfactual DPO
Paper • 2401.09566 • Published • 2 -
BitDelta: Your Fine-Tune May Only Be Worth One Bit
Paper • 2402.10193 • Published • 19 -
Instruction-tuned Language Models are Better Knowledge Learners
Paper • 2402.12847 • Published • 25 -
V-STaR: Training Verifiers for Self-Taught Reasoners
Paper • 2402.06457 • Published • 9 -
Adding NVMe SSDs to Enable and Accelerate 100B Model Fine-tuning on a Single GPU
Paper • 2403.06504 • Published • 53 -
Language models scale reliably with over-training and on downstream tasks
Paper • 2403.08540 • Published • 14 -
LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models
Paper • 2403.13372 • Published • 62 -
RAFT: Adapting Language Model to Domain Specific RAG
Paper • 2403.10131 • Published • 67 -
Quiet-STaR: Language Models Can Teach Themselves to Think Before Speaking
Paper • 2403.09629 • Published • 75 -
Simple and Scalable Strategies to Continually Pre-train Large Language Models
Paper • 2403.08763 • Published • 49 -
Gemma: Open Models Based on Gemini Research and Technology
Paper • 2403.08295 • Published • 47 -
Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context
Paper • 2403.05530 • Published • 61 -
Teaching Large Language Models to Reason with Reinforcement Learning
Paper • 2403.04642 • Published • 46 -
GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection
Paper • 2403.03507 • Published • 183 -
MathScale: Scaling Instruction Tuning for Mathematical Reasoning
Paper • 2403.02884 • Published • 15 -
The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits
Paper • 2402.17764 • Published • 604 -
Nemotron-4 15B Technical Report
Paper • 2402.16819 • Published • 42 -
MegaScale: Scaling Large Language Model Training to More Than 10,000 GPUs
Paper • 2402.15627 • Published • 34 -
Tele-FLM Technical Report
Paper • 2404.16645 • Published • 17 -
The Road Less Scheduled
Paper • 2405.15682 • Published • 21 -
Xwin-LM: Strong and Scalable Alignment Practice for LLMs
Paper • 2405.20335 • Published • 17 -
Perplexed by Perplexity: Perplexity-Based Data Pruning With Small Reference Models
Paper • 2405.20541 • Published • 21 -
Q-GaLore: Quantized GaLore with INT4 Projection and Layer-Adaptive Low-Rank Gradients
Paper • 2407.08296 • Published • 31 -
Power Scheduler: A Batch Size and Token Number Agnostic Learning Rate Scheduler
Paper • 2408.13359 • Published • 22