ModelPrésentéExterne

Hardtests Sft Qwen3.5 27b R16 Epoch1

LoRA adapter from supervised fine-tuning of Qwen/Qwen3.5-27B on the ZhuofengLi/hardtests-sft-qwen3.5-27b dataset (reward=1 competitive-programming trajectories with reasoning).

Sourcehuggingface.co/ZhuofengLi/hardtests-sft-qwen3.5-27b-r16-epoch1/tree/069fc1a51938cd2e599c6cc9b84d559cb4d89c1f

TypeLoRAModèle de baseQwen/Qwen3.5-27BVersionadapter_modelCClasse

Source du modèle

Type: LoRA
Taille du fichier: 223 MB
Modèle de base: Qwen/Qwen3.5-27B
Version: vadapter_model
Mis à jour: 26 sept. 2026
Téléchargements: 8
Source: Hugging Face

Description de la source

HardTests SFT — Qwen3.5-27B LoRA (epoch 1)

LoRA adapter from supervised fine-tuning of Qwen/Qwen3.5-27B on the ZhuofengLi/hardtests-sft-qwen3.5-27b dataset (reward=1 competitive-programming trajectories with reasoning).

Checkpoint: step 100 (recorded epoch = 1.1249112845990064)
LoRA: r=8, alpha=32 (see adapter_config.json)
Contents: adapter weights + configs only (no optimizer state)

Merge into the base model

from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

base = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3.5-27B", torch_dtype="bfloat16", device_map="auto")
model = PeftModel.from_pretrained(base, "ZhuofengLi/hardtests-sft-qwen3.5-27b-r16-epoch1")
model = model.merge_and_unload()
model.save_pretrained("./qwen3.5-27b-hardtests-sft-epoch1-merged")
AutoTokenizer.from_pretrained("Qwen/Qwen3.5-27B").save_pretrained("./qwen3.5-27b-hardtests-sft-epoch1-merged")

Sources

1 source

Hugging FacePrincipalVersion 069fc1a51938cd2e599c6cc9b84d559cb4d89c1f

huggingface.co/ZhuofengLi/hardtests-sft-qwen3.5-27b-r16-epoch1/tree/069fc1a51938cd2e599c6cc9b84d559cb4d89c1f

Vérifié 7 sept.

Artefacts du modèle

1 artefact

adapter_model.safetensors

safetensors · 223 MB · SHA-256 21a9633d8e0c…a95e · Hugging Face

Download file

Extraits de sources

2 extraits

Readme7 sept.huggingface.co/ZhuofengLi/hardtests-sft-qwen3.5-27b-r16-epoch1

--- base_model: Qwen/Qwen3.5-27B library_name: peft tags: - lora - sft - qwen3.5 - hardtests - competitive-programming --- # HardTests SFT — Qwen3.5-27B LoRA (epoch 1) LoRA adapter from supervised fine-tuning of `Qwen/Qwen3.5-27B` on the [ZhuofengLi/hardtests-sft-qwen3.5-27b](https://huggingface.co/datasets/ZhuofengLi/hardtests-sft-qwen3.5-27b) dataset (reward=1 competitive-programming trajectories with reasoning). - **Checkpoint**: step 100 (recorded epoch = 1.1249112845990064) - **LoRA**: r=8, alpha=32 (see `adapter_config.json`) - **Contents**: adapter weights + configs only (no optimizer state) ## Merge into the base model ```python from transformers import AutoModelForCausalLM, AutoTokenizer from peft import PeftModel base = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3.5-27B", torch_dtype="bfloat16", device_map="auto") model = PeftModel.from_pretrained(base, "ZhuofengLi/hardtests-sft-qwen3.5-27b-r16-epoch1") model = model.merge_and_unload() model.save_pretrained("./qwen3.5-27b-hardtests-sft-epoch1-merged") AutoTokenizer.from_pretrained("Qwen/Qwen3.5-27B").save_pretrained("./qwen3.5-27b-hardtests-sft-epoch1-merged") ```

Note7 sept.huggingface.co/ZhuofengLi/hardtests-sft-qwen3.5-27b-r16-epoch1

Source context: 8 downloads · 0 likes · Library peft · Repo ZhuofengLi/hardtests-sft-qwen3.5-27b-r16-epoch1