This repository contains a historical Qwen2.5-7B CoLAR checkpoint exported to a Hugging Face-compatible layout. The repo root works for standard decoding (Transformers/vLLM), and extra_state.pt preserves the CoLAR latent head for latent decoding.

What Is In This Repo

Standard HF model files at repo root (config.json, model*.safetensors, tokenizer files)
extra_state.pt with the latent policy and embedding weights required for CoLAR latent decoding
export_meta.json from the local export step
latent_metadata.json with structured provenance and hyperparameters
artifacts/source_wandb_config.yaml, artifacts/source_wandb_summary.json, and artifacts/source_wandb_metadata.json
artifacts/source_all_config.json parsed from the original WandB all_config

Source Provenance

Field	Value
WandB run	`b7fyxsug`
Run date	`2025-09-29`
Task	Narrative continuation planning
Stage	post-SFT
Base model	`Qwen/Qwen2.5-7B-Instruct`
Git commit	`3709e7a067a6c6e033942b768d6cc88ba54e82d0`
Host / GPU	`alexgurung-latent-qcvnm-ctgn8` / `NVIDIA H200`
Original checkpoint	`/mnt/volume3/baseline_colar/colar/oldlogs/colar/colar-experiments/b7fyxsug/checkpoints/pt_epoch1__step451__monitor3.139.ckpt`
Generalization slug	`colar_ncp_post_sft`

Usage

Standard decoding

python

from transformers import AutoModelForCausalLM, AutoTokenizer

repo_dir = "agurung/colar-qwen25-7b-ncp-post-sft"
model = AutoModelForCausalLM.from_pretrained(repo_dir, torch_dtype="auto", device_map="auto")
tokenizer = AutoTokenizer.from_pretrained(repo_dir)

Latent decoding

python

from huggingface_hub import snapshot_download
from pathlib import Path

local_dir = Path(snapshot_download("agurung/colar-qwen25-7b-ncp-post-sft"))
llm_dir = local_dir
extra_state = local_dir / "extra_state.pt"

Example evaluator invocation:

python -m litereason.experiments.latent_eval.colar.evaluate_colar   --llm-dir "$LOCAL_DIR"   --extra-state "$LOCAL_DIR/extra_state.pt"   --test-file litereason/experiments/generalization/data/gsm8k.jsonl   --prompt-variant standard   --save-preds preds_gsm8k_standard.jsonl   --num-samples 5   --use-chat-template

Key Hyperparameters

Field	Value
Trainer strategy	`deepspeed_stage_2_offload`
Max epochs	`2`
Precision	`bf16-mixed`
Global batch size	`1`
Validation batch size	`4`
Compression factor	`5`
Max latent forward	`64`
Latent temperature	`1.0`
Group size	`8`
Experience batch size	`64`
Train samples / epoch	`512`

Reported Metrics

Metric	Value
`monitor`	`3.1386220455169678`
`val/loss`	`3.1386220455169678`
`val/ce_loss`	`2.084620952606201`
`val/embed_modeling_loss`	`1.0540010929107666`
`epoch`	`1`
`trainer/global_step`	`450`

Notes

Historical CoLAR NCP supervised fine-tuning checkpoint.
Validation loss in the checkpoint filename is monitor=3.139.

Local Reference Paths

WandB config: /mnt/volume3/baseline_colar/colar/oldlogs/colar/wandb/run-20250929_200617-b7fyxsug/files/config.yaml
WandB summary: /mnt/volume3/baseline_colar/colar/oldlogs/colar/wandb/run-20250929_200617-b7fyxsug/files/wandb-summary.json
WandB metadata: /mnt/volume3/baseline_colar/colar/oldlogs/colar/wandb/run-20250929_200617-b7fyxsug/files/wandb-metadata.json
Local export dir: /mnt/disk/baseline_colar/colar/exported_hf/b7fyxsug