cmndcntrlcyber

gemma4-26b-a4b-dapt-offsec

README

License: apache-2.0

Model architecture notes

Gemma 4 26B-A4B is a Mixture-of-Experts model: 128 experts + 1 shared expert, 8 active per layer, 30 layers (25 sliding-window @ 1024 + 5 full attention). Total parameters: 25.8B; active per forward pass: 3.8B.

LoRA targeting constraint: the routed expert FFN layers use 3D nn.Parameter tensors that PEFT cannot target. LoRA is applied only to shared attention + shared MLP modules (q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj). Domain vocabulary is still injected through these shared pathways.

Gemma4ClippableLinear: Gemma 4 wraps linear layers in Gemma4ClippableLinear modules that must be unwrapped before PEFT operations. This is handled by src/utils.py:unwrap_clippable_linear.

Intended use

Direct use: not recommended — this is a domain-adaptation adapter, not instruction-tuned beyond the base model's existing capabilities.
Downstream: merge into the base model before SFT training. The Gemma pipeline merges this adapter first, then trains the SFT adapter (gemma4-26b-a4b-code-trainer-aggressive-full1) on the merged weights.
Out of scope: general-purpose chat, safety-critical applications, or non-code tasks.

Training data

Dataset: cmndcntrlcyber/gemma4-26b-a4b-dapt-offsec-corpus
Rows: 20,680 offensive security code documents
Format: {"text": "..."} — plain text continuation (no chat formatting)
Source: offensive-security GitHub repositories, cloned and chunked by src/phase3b_dapt/data/prepare_corpus.py
Shared with: the Qwen DAPT pipeline uses the same underlying corpus

Training procedure

Table with columns: Knob, Value
Knob	Value
Base model	`google/gemma-4-26B-A4B-it`
Adapter	LoRA (PEFT), `r = 32`, `alpha = 64`, `dropout = 0.05`
Target modules	`q_proj`, `k_proj`, `v_proj`, `o_proj`, `gate_proj`, `up_proj`, `down_proj`

Table with columns: Meta, Value
Meta	Value
Hardware	HF Jobs `a100-large` (1x A100 80 GB)
Entry point	`src/phase3b_dapt/hf_skills/dapt_entry.py`
Config	`src/config/pipeline-gemma26b.yml` (`gemma_dapt` section)

Limitations

Shared layers only. LoRA cannot target routed expert FFN (3D nn.Parameter), so domain adaptation is limited to shared attention and shared MLP pathways.
No safety tuning. Inherits the base model's safety properties.
Not for direct use. This adapter is a pipeline intermediate — merge it before SFT for best results.

How to use

python

import torchfrom peft import PeftModelfrom transformers import AutoModelForCausalLM, AutoTokenizer base_id = "google/gemma-4-26B-A4B-it"dapt_id = "cmndcntrlcyber/gemma4-26b-a4b-dapt-offsec" tokenizer = AutoTokenizer.from_pretrained(base_id)model = AutoModelForCausalLM.from_pretrained(    base_id, torch_dtype=torch.bfloat16, device_map="auto",) # Unwrap Gemma4ClippableLinear before PEFT operationsfrom src.utils import unwrap_clippable_linearunwrap_clippable_linear(model) model = PeftModel.from_pretrained(model, dapt_id)model = model.merge_and_unload()  # merge into base for downstream SFT

Reproducibility

Code: github.com/cmndcntrlcyber/code-trainer-pipeline (src/phase3b_dapt/)

Corpus build:

bash

python -m src.phase3b_dapt.data.prepare_corpus \    --config src/config/pipeline-gemma26b.yml

Training launch:

bash

python -m src.phase3b_dapt.scripts.launch_dapt \    --config src/config/pipeline-gemma26b.yml --wait

Available on FriendliAI

Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more

Container

Run this model inference with full control and performance in your environment.

Learn more

Model Details

Model Provider

cmndcntrlcyber

Model Tree

Base

google/gemma-4-26B-A4B-it

Adapter

this model

Input Modalities

TextImage

Output Modalities

Text

Supported Functionality

Dedicated EndpointsContainer

Explore FriendliAI today

Get started Talk to an engineer

README

License: apache-2.0

Model architecture notes

Gemma4ClippableLinear: Gemma 4 wraps linear layers in Gemma4ClippableLinear modules that must be unwrapped before PEFT operations. This is handled by src/utils.py:unwrap_clippable_linear.

Intended use

Direct use: not recommended — this is a domain-adaptation adapter, not instruction-tuned beyond the base model's existing capabilities.
Downstream: merge into the base model before SFT training. The Gemma pipeline merges this adapter first, then trains the SFT adapter (gemma4-26b-a4b-code-trainer-aggressive-full1) on the merged weights.
Out of scope: general-purpose chat, safety-critical applications, or non-code tasks.

Training data

Dataset: cmndcntrlcyber/gemma4-26b-a4b-dapt-offsec-corpus
Rows: 20,680 offensive security code documents
Format: {"text": "..."} — plain text continuation (no chat formatting)
Source: offensive-security GitHub repositories, cloned and chunked by src/phase3b_dapt/data/prepare_corpus.py
Shared with: the Qwen DAPT pipeline uses the same underlying corpus

Training procedure

Table with columns: Knob, Value
Knob	Value
Base model	`google/gemma-4-26B-A4B-it`
Adapter	LoRA (PEFT), `r = 32`, `alpha = 64`, `dropout = 0.05`
Target modules	`q_proj`, `k_proj`, `v_proj`, `o_proj`, `gate_proj`, `up_proj`, `down_proj`

Table with columns: Meta, Value
Meta	Value
Hardware	HF Jobs `a100-large` (1x A100 80 GB)
Entry point	`src/phase3b_dapt/hf_skills/dapt_entry.py`
Config	`src/config/pipeline-gemma26b.yml` (`gemma_dapt` section)

Limitations

Shared layers only. LoRA cannot target routed expert FFN (3D nn.Parameter), so domain adaptation is limited to shared attention and shared MLP pathways.
No safety tuning. Inherits the base model's safety properties.
Not for direct use. This adapter is a pipeline intermediate — merge it before SFT for best results.

How to use

python

import torchfrom peft import PeftModelfrom transformers import AutoModelForCausalLM, AutoTokenizer base_id = "google/gemma-4-26B-A4B-it"dapt_id = "cmndcntrlcyber/gemma4-26b-a4b-dapt-offsec" tokenizer = AutoTokenizer.from_pretrained(base_id)model = AutoModelForCausalLM.from_pretrained(    base_id, torch_dtype=torch.bfloat16, device_map="auto",) # Unwrap Gemma4ClippableLinear before PEFT operationsfrom src.utils import unwrap_clippable_linearunwrap_clippable_linear(model) model = PeftModel.from_pretrained(model, dapt_id)model = model.merge_and_unload()  # merge into base for downstream SFT

import torchfrom peft import PeftModelfrom transformers import AutoModelForCausalLM, AutoTokenizer base_id = "google/gemma-4-26B-A4B-it"dapt_id = "cmndcntrlcyber/gemma4-26b-a4b-dapt-offsec" tokenizer = AutoTokenizer.from_pretrained(base_id)model = AutoModelForCausalLM.from_pretrained(    base_id, torch_dtype=torch.bfloat16, device_map="auto",) # Unwrap Gemma4ClippableLinear before PEFT operationsfrom src.utils import unwrap_clippable_linearunwrap_clippable_linear(model) model = PeftModel.from_pretrained(model, dapt_id)model = model.merge_and_unload()  # merge into base for downstream SFT

Reproducibility

Code: github.com/cmndcntrlcyber/code-trainer-pipeline (src/phase3b_dapt/)

Corpus build:

bash

python -m src.phase3b_dapt.data.prepare_corpus \    --config src/config/pipeline-gemma26b.yml

python -m src.phase3b_dapt.data.prepare_corpus \    --config src/config/pipeline-gemma26b.yml

Training launch:

bash

python -m src.phase3b_dapt.scripts.launch_dapt \    --config src/config/pipeline-gemma26b.yml --wait

python -m src.phase3b_dapt.scripts.launch_dapt \    --config src/config/pipeline-gemma26b.yml --wait