method-mtma-0825-0119

agentic-moral-alignment/method-mtma-0825-0119

method-mtma-0825-0119 is a machine learning model from agentic-moral-alignment. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft.

LoRA adapters from mt-moral-alignment run method-mtma-0825-0119, one subfolder per checkpoint, converted from verl FSDP shards. Per-episode logs: https://huggingface.co/datasets/agentic-moral-alignment/mtma/tree/main/…

Downloads · 30 days

Access

Public

Updated Aug 25, 2026

Repo size

1 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors1 GB · 100%

At a glance

Library: peft
Access: Public
Created: Aug 25, 2026
Updated: Aug 25, 2026
SHA: fe84672d

Base models

Qwen/Qwen3.5-4B-Base

Model card

From the Hugging Face model README

method-mtma-0825-0119

LoRA adapters from mt-moral-alignment run method-mtma-0825-0119, one subfolder per checkpoint, converted from verl FSDP shards. Per-episode logs: https://huggingface.co/datasets/agentic-moral-alignment/mtma/tree/main/runs/train/method-mtma-0825-0119

step	load
5	`subfolder="step_5"`
10	`subfolder="step_10"`
15	`subfolder="step_15"`
20	`subfolder="step_20"`
25	`subfolder="step_25"`
30	`subfolder="step_30"`
35	`subfolder="step_35"`
40	`subfolder="step_40"`

base = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3.5-4B-Base")
model = PeftModel.from_pretrained(base, "agentic-moral-alignment/method-mtma-0825-0119", subfolder="step_40")

WARNING: vLLM cannot serve these adapters on Qwen3.5 -- hybrid-GDN LoRA is a silent no-op (vllm#49354). Merge first: model.merge_and_unload().save_pretrained(...).

Frozen run config

# frozen for method-mtma-0825-0119 (2026-08-25T08:10:15Z, git 09dfee3)
STAGE: train
MODEL: Qwen/Qwen3.5-4B-Base
GPUS: 1
TP: 1
MAX_SEQS: 64
GPU_MEM_UTIL: 0.72
GAMES: neighbourhood-necromancer
BATCH_GAME: 4
START_SAMPLE: True
START_RESAMPLE_STEP: True
TURN_CAP: 64
N: 8
STEPS: 40
LR: 1.5e-05
TOP_P: 0.95
ROLLOUT_SEED: 42
FRAMING: game-goals
LAM: 1
TASK_WEIGHT: 0
MORAL: deon_soft
RETURN_FN: reward_to_go
BASELINE_FN: state_mean
NORMALIZE_FN: none
GAMMA: 0.0
OMEGA: 0
TD_LAMBDA: 0.8
THINKING: True
MAX_TURN_TOKENS: 4096
ENV_TIMEOUT_S: 60
AGENT_WORKERS: 2
REWARD_WORKERS: 1
BYPASS: True
GRAD_CKPT: True
CALC_ENTROPY: True
SAVE_FREQ: 5
NOSET: 0
PUSH: 0
PROMPT_LEN: 4096
RESPONSE_LEN: 40960
MAX_MODEL_LEN: 45056
TOKEN_LEN_PER_GPU: 49152
LORA_RANK: 32
LORA_ALPHA: 64

Library: peft
Created: Aug 25, 2026
Updated: Aug 25, 2026