You can deploy this model and start earning money today!

rjz123/
colar-strategyqa-l1b

Name: colar-strategyqa-l1b
Availability: InStock

unsloth/Llama-3.2-1B-Instruct

finetune

llama3.2

Views

No views yet

CoLaR — StrategyQA (commonsense), Llama-3.2-1B

A CoLaR (Compressed Latent Reasoning) checkpoint for Commonsense (StrategyQA), fine-tuned from unsloth/Llama-3.2-1B-Instruct. The model reasons in compressed continuous latent embeddings rather than explicit chain-of-thought tokens.

Model

Base model: unsloth/Llama-3.2-1B-Instruct
Framework: CoLaR — Think Silently, Think Fast: Dynamic Latent Compression of LLM Reasoning Chains (arXiv:2505.16552).
Domain: Commonsense (StrategyQA)
Warm-start: self-trained CoT baseline (bundled cot_baseline.ckpt)
Files: cot_baseline.ckpt, sft_adaptiveLRM.ckpt

Training procedure

Trained with the CoLaR supervised fine-tuning (SFT) recipe: the frozen base LLM is adapted with q/v LoRA (rank 128, alpha 32) plus a trainable Latent Head (3-layer MLP) and an embedding-compression module. The objective is next-token cross-entropy on the answer plus an embed_modeling_loss (MSE) that reconstructs the compressed reasoning-step embeddings (each latent token summarizes about compression_factor chain-of-thought tokens). Reinforcement learning (GRPO) is disabled — this is an SFT-only checkpoint.

This checkpoint: Self-trained: CoT baseline + CoLaR-SFT adaptive-LRM (RL off). Bundle includes cot_baseline.ckpt (CoT stage) and sft_adaptiveLRM.ckpt (latent stage).

Datasets

StrategyQA

How to load

This is a PyTorch-Lightning checkpoint (weights under the top-level key state_dict) that fits the CoLaR scaffold — it is not directly AutoModel-loadable. Load the base model, splice this state_dict in with strict=False, and use the CoLaR runtime settings:

COLAR_EMB_STD=0.018   COLAR_COMPRESS=<compression_factor>   sep_token=###
TORCH_FORCE_NO_WEIGHTS_ONLY_LOAD=1

See the CoLaR repository (github.com/xiaomi-research/colar) for the exact loader. The shared GSM8K warm-start ancestor is the official AlbertTan/CoLaR release.

Research artifact for latent-reasoning study (small 1–1.5B model).

rjz123/colar-strategyqa-l1b