Beta

Explore
Marketplace
Neural Labs
Chat
Wallet
Docs

Beta

Explore Marketplace Neural Labs Wallet Docs

Build

Manage

My assets Manage wallet Organization

Community

Discord X (Twitter)Telegram

Docs Privacy Policy Terms and Conditions

Proud member of

Proud member of

Docs Cookie Policy

Privacy Policy Terms and Conditions

© 2026 AlphaNeural AI. All rights reserved.

gemma-4-e2b-rft-mutation-merged – AI Model by bookxd | AlphaNeural AI

You can deploy this model and start earning money today!

bookxd/
gemma-4-e2b-rft-mutation-merged

like

0

image-text-to-text

rejection-fine-tuning

mutation-testing

text-generation

google/gemma-4-E2B-it

endpoints_compatible

Views

No views yet

Gemma 4 E2B — merged RFT weights (mutation corpus)

Full merged weights (LoRA baked into base) from rejection fine-tuning on the merged mutation corpus (Commons Lang + fastutil, 56 train samples).

This is a standalone model — load directly without PEFT:

python

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_id = "bookxd/gemma-4-e2b-rft-mutation-merged"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.bfloat16,
    attn_implementation="sdpa",
    device_map="auto",
)

LoRA adapter (unmerged): bookxd/gemma-4-e2b-rft-mutation

Merged with the project KV-share-safe state-dict merger (205 LoRA targets, scaling=2.0).