Modelpublic

phuctahong/nanogpt-falcon-1-3-lambda-init-0-1-0-5-lambda-scale-0-7-job7990408-step100000

sourceHugging Faceapache-2.0updated 1mo agoView on Hugging Face

0likes41downloads

Model Card

phuctahong/nanogpt-falcon-1-3-lambda-init-0-1-0-5-lambda-scale-0-7-job7990408-step100000

NanoGPT Pro checkpoint uploaded from the princeton-pli/Nanogpt W&B project.

Provenance

—W&B run: [[Falcon-1.3-lambda-init-0.1-0.5-lambdascale-0.7]Mgpt-mha-scaleddeltanetparallelctxetactxlambdatriton130mOptadamwLR0.001Dfineweb-edu100BT49.15Btime20260512105338jobid7990408](https://wandb.ai/princeton-pli/Nanogpt/runs/5xfob5zq)
—W&B run id: 5xfob5zq
—Job id: 7990408
—Checkpoint step: 100000
—Local checkpoint: /scratch/gpfs/ARORA/st0490/nanogpt-next_mod/output/out_gpt-mha-scaled_delta_net_parallel_ctxeta_ctxlambda_triton_130m_Opt_adamw_LR_0.001_D_fineweb-edu100B_T_49.15B_time_20260512_105338_jobid_7990408/checkpoint-100000

Files

This repo contains inference-ready weights only:

—config.json
—model.safetensors
—generation_config.json

Optimizer and trainer state were intentionally omitted.

Load locally

python

from nanogptpro.model_loader import load_model_from_checkpoint

model = load_model_from_checkpoint(
    model_id="<your-model-id>",
    checkpoint="phuctahong/nanogpt-falcon-1-3-lambda-init-0-1-0-5-lambda-scale-0-7-job7990408-step100000",
)

Install nanogptpro and use the model id that matches the architecture recorded in config.json.