Metadata-Version: 2.4
Name: brainpatch
Version: 1.2.0
Summary: A portable format and runtime for reversible activation-space interventions in frozen language models.
Author: Blackdrome AI Labs
Maintainer-email: Blackdrome AI Labs <hello@blackdrome.tech>
License: Apache-2.0
Project-URL: Homepage, https://github.com/blackdromeai-labs/brainpatch
Project-URL: Repository, https://github.com/blackdromeai-labs/brainpatch
Project-URL: Documentation, https://github.com/blackdromeai-labs/brainpatch/tree/main/docs
Project-URL: Issues, https://github.com/blackdromeai-labs/brainpatch/issues
Project-URL: Changelog, https://github.com/blackdromeai-labs/brainpatch/blob/main/CHANGELOG.md
Project-URL: Organization, https://github.com/blackdromeai-labs
Keywords: activation-steering,interpretability,llm,mechanistic-interpretability,activation-patching,model-steering,inference-time-intervention
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: typer>=0.12
Requires-Dist: rich>=13.0
Provides-Extra: transformers
Requires-Dist: torch>=2.1; extra == "transformers"
Requires-Dist: transformers>=4.40; extra == "transformers"
Requires-Dist: accelerate>=0.30; extra == "transformers"
Provides-Extra: vllm
Requires-Dist: vllm>=0.6; extra == "vllm"
Requires-Dist: torch>=2.1; extra == "vllm"
Provides-Extra: mlx
Requires-Dist: mlx-lm>=0.20; (platform_system == "Darwin" and platform_machine == "arm64") and extra == "mlx"
Provides-Extra: llamacpp
Requires-Dist: gguf>=0.10; extra == "llamacpp"
Requires-Dist: numpy>=1.24; extra == "llamacpp"
Provides-Extra: hub
Requires-Dist: huggingface_hub>=0.25; extra == "hub"
Provides-Extra: server
Requires-Dist: fastapi>=0.110; extra == "server"
Requires-Dist: uvicorn>=0.30; extra == "server"
Requires-Dist: pydantic>=2.0; extra == "server"
Provides-Extra: ui
Requires-Dist: gradio>=4.44; extra == "ui"
Provides-Extra: research
Requires-Dist: torch>=2.1; extra == "research"
Requires-Dist: transformers>=4.40; extra == "research"
Requires-Dist: datasets>=2.19; extra == "research"
Requires-Dist: accelerate>=0.30; extra == "research"
Requires-Dist: safetensors>=0.4; extra == "research"
Requires-Dist: numpy>=1.24; extra == "research"
Requires-Dist: pyyaml>=6.0; extra == "research"
Requires-Dist: huggingface_hub>=0.25; extra == "research"
Provides-Extra: modal
Requires-Dist: modal>=1.5.0; extra == "modal"
Provides-Extra: all
Requires-Dist: brainpatch[hub,llamacpp,server,transformers,ui]; extra == "all"
Provides-Extra: dev
Requires-Dist: pytest>=8.0; extra == "dev"
Requires-Dist: build>=1.2; extra == "dev"
Dynamic: license-file

# BrainPatch

**A portable format and runtime for reversible activation-space interventions in
frozen language models.**

No fine-tuning. No weight modification. No prompt injection. No hosted service.

An open-source system from [Blackdrome AI Labs](https://blackdrome.tech).

```bash
pip install "brainpatch[transformers]"
brainpatch install ./experimental-feature-727.brainpatch
brainpatch compare --model Qwen/Qwen2.5-1.5B-Instruct \
                   --patch experimental-feature-727 \
                   --prompt "Explain why the sky is blue."
```

```python
from brainpatch import BrainPatchedModel

model = BrainPatchedModel.from_pretrained(
    "Qwen/Qwen2.5-1.5B-Instruct", backend="transformers", device="auto"
)
patch = model.install("./experimental-feature-727.brainpatch")
patch.strength = 0.8
print(model.generate("Evaluate my idea."))
```

[![License](https://img.shields.io/badge/license-Apache--2.0-blue.svg)](https://github.com/blackdromeai-labs/brainpatch/blob/main/LICENSE)
[![Format](https://img.shields.io/badge/format-.brainpatch%20v1-orange.svg)](https://github.com/blackdromeai-labs/brainpatch/blob/main/docs/patch-format.md)
[![Blackdrome AI Labs](https://img.shields.io/badge/Blackdrome%20AI%20Labs-open%20source-black.svg)](https://github.com/blackdromeai-labs)

---

## What it is

A **BrainPatch** is a small file containing one or more direction vectors and the
layers to add them to. Loading one installs a behavioural intervention into a
model whose weights are never touched.

BrainPatch is **infrastructure, not a method**. It is agnostic about how a
direction was discovered — difference-of-means, PCA, a linear probe, an SAE
feature, or anything else — and it is not tied to any one behaviour. The project
began as an SAE tool; on our own benchmark the SAE finished behind PCA and
difference-of-means, and the format outlived that assumption.

| | LoRA / fine-tune | Prompt engineering | **BrainPatch** |
|---|---|---|---|
| changes weights | yes | no | **no** |
| artifact size | MB – GB | n/a | **6.7 KB** (measured, below) |
| costs context | no | yes, every call | **no** |
| adjustable at runtime | no | crudely | **yes, continuously** |
| removable mid-session | no | yes | **yes** |
| changeable *during* generation | no | no | **yes** (token schedules) |

Measured on the reference patch in this repository: **6,856 bytes**, against
**3,087,467,144 bytes** of Qwen2.5-1.5B-Instruct weights — a ratio of
**450,331×**. Runtime overhead on an L4 was **−1.3% (within noise)** on
Transformers and **+2.2% (tokens/second)** on vLLM, with
**0.01 MB** of extra VRAM.

## Read this before using the example patches

The runtime is verified. **The example patches' behaviour is weakly established
and we label them accordingly.** BrainPatch will not name a patch after a
behaviour it has not demonstrated.

| patch | evidence | what it is |
|---|---|---|
| `experimental-feature-727` | `none` | named after its feature ID because its controls came back negative — a scale-matched random direction moved the output *further* from baseline than the real one. Ships as a working demonstration of the format. |
| `experimental-commitment` | `controlled_interventional` | a real, replicated effect on commitment / anti-evasion. **Formerly named `anti-sycophancy`; that name was withdrawn** because sycophantic agreement *rose* under it. |
| `experimental-independent-criticism-candidate` | `none` | a candidate direction, never validated. |

## What the research established, in order

Four experiments, four times the conclusion changed. The write-ups include the
numbers that looked good before they didn't.

**`anti_sycophancy_v1` — negative.** 198 propositions, topic-disjoint splits,
true-assertion controls so "disagree with everything" cannot score as
independence, criteria [pre-registered](https://github.com/blackdromeai-labs/brainpatch/blob/main/experiments/anti_sycophancy_v1/success_criteria.md)
before the test split was opened. The selected direction beat all ten
scale-matched random directions, beat three unrelated real directions and
reversed under sign inversion — but its per-item effect correlated **+0.457**
with response-length gap, over a threshold set in advance, and free generation
moved the wrong way. [Full write-up](https://github.com/blackdromeai-labs/brainpatch/blob/main/experiments/anti_sycophancy_v1/RESULTS.md).

Three findings outlived the null:

- **The SAE came last.** Both SAE variants lost to PCA, to a linear probe and to difference-of-means, and both failed the true-claim control by making the model disagree with *true* statements.
- **Probe accuracy is not steerability.** A probe separating the classes at **100%** accuracy steered worse than PCA, which uses no labels when fitting. Readable and pushable are different properties.
- **Where you inject beats almost everything else.** Prompt-token steering was ~**6×** more effective than steering generated tokens.

**`anti_sycophancy_v2` — negative, test split never opened.** Rebuilt the dataset
to remove the length confound (preferred response longer in **53%** of pairs
instead of 96%). The pre-registered free-generation gate stopped it at
validation. It produced the most useful finding in the project, across 27
configurations that passed every log-probability gate:

```
corr(log-probability effect, free-generation correction gain) = −0.298
```

**Ranking activation-steering directions by paired log-probability
*anti-selects* for the behaviour you want in generation.** That explains v1
exactly: its winner had a strong, control-beating log-prob effect and a *falling*
correction rate. Only **27 of 330** configurations survived the true-claim guard —
most directions that "work" are simply contrarian.
[Full write-up](https://github.com/blackdromeai-labs/brainpatch/tree/main/experiments/anti_sycophancy_v2).

**`anti_sycophancy_v3` — the first positive result.** Selecting by *generated
behaviour* instead of log-probability, on a third fresh dataset, a single SAE
feature injected at the prompt raised the correction rate on false user
assertions from **0.233 to 0.400** on 200 held-out items — CI [+0.092, +0.242],
McNemar **p = 3.6 × 10⁻⁵**, all **11** pre-registered gates passed.
[Full write-up](https://github.com/blackdromeai-labs/brainpatch/tree/main/experiments/anti_sycophancy_v3).

**`anti_sycophancy_specificity_v1` — replicated the effect, then took its name
away.** On 407 fresh propositions with zero overlap against all prior data, the
effect reproduced *through the compiled artifact*: correction **0.248 → 0.376**,
CI [+0.072, +0.184], McNemar **p = 2.1 × 10⁻⁵**.

Against **100** norm-matched random directions it reached the **95.0th
percentile**, empirical **p = 0.0594** — a pre-registered pass, but 5 of 100
random directions matched or beat it. Specificity is real and weak.

Then the transition matrix showed what the headline hid. Only ~**22%** of the
gain (+0.028, p = 0.17) was a claim moving from *agreement* to *challenge*. Most
was evasive replies becoming committed ones, landing correct or sycophantic at
chance (28 vs 26, p = 0.89) — and **sycophantic agreement rose, 0.280 → 0.352**.
Removing the user's endorsement kept 75% of the effect.

So the patch was renamed **`experimental-commitment`**. It steers commitment and
suppresses evasion; it does not reduce sycophancy. The evidence level was
deliberately *not* upgraded ([why](https://github.com/blackdromeai-labs/brainpatch/blob/main/docs/evidence_level_decision.md)).
[Full write-up](https://github.com/blackdromeai-labs/brainpatch/tree/main/experiments/anti_sycophancy_specificity_v1).

**`artifact_fidelity_v1` — a claim of ours, refuted.** We had shipped an artifact
with an inverted coefficient sign that passed a cosine check at 1.0, and
hypothesised that behavioural testing was therefore *required* to catch such
defects. Thirteen controlled corruptions said otherwise: signed cosine, delta
norm and execution tracing together caught **all eleven** real defects, and
behavioural reproduction caught **nothing they missed** while missing two they
caught. The lesson was narrower than the hypothesis — *the reference had been
unsigned* — and it is now the [`brainpatch verify`](https://github.com/blackdromeai-labs/brainpatch/blob/main/docs/verification.md)
workflow. [Full write-up](https://github.com/blackdromeai-labs/brainpatch/blob/main/experiments/artifact_fidelity_v1/report.md).

Total compute for all five experiments: **~$4.65**. None of these failures needed
scale to find; they needed the next question to be asked.

---

## Install

```bash
pip install brainpatch                    # core: format, registry, CLI. No ML stack.
pip install "brainpatch[transformers]"    # PyTorch backend (CUDA / CPU / MPS)
pip install "brainpatch[llamacpp]"        # GGUF control-vector export
pip install "brainpatch[vllm]"            # high-throughput serving
pip install "brainpatch[server]"          # OpenAI-compatible HTTP API
pip install "brainpatch[ui]"              # local web UI
pip install "brainpatch[research]"        # patch authoring: SAEs, extraction
```

The core wheel pulls in **only `typer` and `rich`** — no torch, no numpy, not
even `safetensors` (the container is parsed by a pure-Python reader). Installing,
inspecting and validating a patch works on a bare Python 3.10+.

```bash
brainpatch doctor      # which engines are installed and usable
brainpatch backends    # full capability matrix
```

## Supported backends

| Backend | Status | Static | Schedules | Server | Verified against |
|---|---|---|---|---|---|
| **Transformers** | **verified** | ✅ | ✅ | ✅ | Qwen2.5-1.5B bf16, NVIDIA L4 |
| **llama.cpp** | **verified** | ✅ | ❌ | ✅ | upstream **b10344**, **Q4_K_M** GGUF |
| **vLLM** | **verified** | ✅ | ❌ | ✅ | **vLLM 0.11.0**, L4, OpenAI server |
| **MLX-LM** | experimental | ✅ | ❌ | ❌ | never run on Apple Silicon |

"Verified" means an automated acceptance suite ran against a real model and
passed. "Implemented" means the adapter is written and reviewed but no hardware
has confirmed it. We do not use the word "supported" for the latter.

What "verified" cost, concretely — each backend ran an automated acceptance
suite against a real model:

- **Transformers**: weights provably unchanged, `strength=0` byte-identical to
  baseline, measured delta norm **28.5177 == expected 28.5177**, schedules fire
  at the keyframe, disable/remove restore baseline.
- **llama.cpp**: 0-based BrainPatch layer 18 maps to the `direction.19` tensor
  (1-based), scale-0 output character-identical to baseline, non-zero scale
  changes output, no crash — on a real 1.12 GB Q4_K_M GGUF.
- **vLLM**: hooks confirmed **inside the worker process** (`Qwen2ForCausalLM`,
  28 layers, `active_hooks: 1`, `cuda:0`), OpenAI server serves concurrent
  requests with no state leak, mismatched per-request strength rejected 400.

Capability gaps are real, not oversights:

- **llama.cpp has no token schedules** — a control vector is bound for a whole
  run and the CLI exposes no per-decode-step control.
- **vLLM has no per-request strength** — continuous batching means one forward
  pass serves many sequences, so a per-request coefficient would change *other
  users'* output. Patch state is frozen while serving, which is what makes
  concurrency safe.
- **Quantization is not assumed to transfer.** A direction fitted on bf16 was
  checked to still change output at Q4_K_M; whether it produces the *same*
  behavioural effect at 4-bit is untested.

## CLI

```bash
brainpatch install <file.brainpatch | owner/repo>
brainpatch list
brainpatch inspect <name>
brainpatch validate <name> --model Qwen/Qwen2.5-1.5B-Instruct
brainpatch verify <name> --reference known-good.brainpatch --model <model>
brainpatch run "prompt" --model <model> --patch <name>
brainpatch compare --model <model> --patch <name> --prompt "..."
brainpatch chat --model <model> --patch <name>
brainpatch serve --model <model> --patch <name> --port 8000
brainpatch ui
brainpatch compile research.json --sae ./sae.pt -o out.brainpatch
brainpatch compile out.brainpatch --backend llama.cpp -o cv.gguf
brainpatch benchmark --model <model> --patch <name>
brainpatch doctor
brainpatch backends
```

## Verifying a patch

`validate` asks *is this file well-formed and loadable*. `verify` asks *does it
do what it says*. Those are different questions:

```bash
brainpatch verify ./my-patch.brainpatch     --reference ./known-good.brainpatch     --model Qwen/Qwen2.5-1.5B-Instruct
```

```
numerical
  signed_direction  FAIL  signed cosine -1.00000000 (threshold 0.999000);
                          unsigned |cos| 1.00000000 would have passed
```

Four levels, each catching what the one above it cannot. A level whose inputs are
missing is reported `skipped`, never dropped — an absent check reads as a pass.

| level | needs | catches |
|---|---|---|
| structural | nothing | corrupt archive, malformed manifest, wrong shape, wrong base model |
| numerical | `--reference` | wrong direction **including sign**, wrong magnitude |
| execution | `--model` | wrong layer, wrong injection site, lost schedule |
| behavioural | a recorded result | drift that survives all of the above |

We built this because we shipped a sign-inverted artifact that passed every
file-level check. Measured across thirteen controlled corruptions: schema,
checksum, shape and model-compatibility caught **0 of 11** real defects; layer and
site defects were invisible to *every* file-level check including signed cosine.
Only tracing the runtime found them.

**Checksums prove a file was not modified. They prove nothing about whether it
encodes the right intervention.** Full detail: [docs/verification.md](https://github.com/blackdromeai-labs/brainpatch/blob/main/docs/verification.md).

## OpenAI-compatible server

```bash
brainpatch serve --model Qwen/Qwen2.5-1.5B-Instruct --patch my-patch --port 8000
```

```python
from openai import OpenAI
client = OpenAI(base_url="http://localhost:8000/v1", api_key="not-needed")
client.chat.completions.create(model="qwen", messages=[{"role": "user", "content": "hi"}])
```

Existing clients work unchanged. Patch strength is configured at **startup**, not
per request — with a shared model, honouring a per-request strength would alter
other in-flight requests' output. A mismatched `brainpatch` extra field returns
a clear 400 rather than being silently ignored.

## The `.brainpatch` format

A ZIP containing only inert data:

```
manifest.json         what to add, where, how strongly
vectors.safetensors   the direction vectors
checksums.json        sha256 of every member
README.md             optional
```

**A patch cannot execute code.** No pickle, no scripts. The loader reads members
by exact name, rejects unexpected members, absolute paths, `..` traversal,
symlinks and zip bombs, and verifies every checksum before use. Archives are
byte-deterministic, so a published patch has a stable hash.

Compatibility is enforced in three modes — `strict` (default; model id and
revision must match), `architecture`, and `unsafe` — because a direction fitted
in one model's basis means nothing in another's.

Full specification: [docs/patch-format.md](https://github.com/blackdromeai-labs/brainpatch/blob/main/docs/patch-format.md).

## Architecture

```mermaid
flowchart TD
    REPO["patch repository<br/>(Hugging Face, a file, anywhere)"]
    ART["tiny .brainpatch<br/>~6 KB"]
    REPO --> ART

    ART --> LOCAL["local workstation"]
    ART --> CLOUD["cloud server"]

    LOCAL --> TF["Transformers"]
    LOCAL --> LC["llama.cpp"]
    LOCAL --> MLX["MLX"]
    CLOUD --> VLLM["vLLM"]

    TF --> FROZEN["frozen base LLM<br/>weights untouched"]
    LC --> FROZEN
    MLX --> FROZEN
    VLLM --> FROZEN

    style ART fill:#fff4e6,stroke:#d99b4a
    style FROZEN fill:#eefaf0,stroke:#4ad97a
```

The runtime knows nothing about where a patch was trained. Separately, the
research toolkit is how patches are *made*:

```mermaid
flowchart TD
    A["activation extraction"] --> B["sparse autoencoder"]
    B --> C["feature discovery"]
    C --> D["causal validation<br/>+ controls"]
    D --> E["patch compile"]
    E --> F["publish"]
    F -.-> G["a .brainpatch file"]

    H["Modal · local GPU · cluster<br/>(optional research backend)"] -.-> A

    style G fill:#fff4e6,stroke:#d99b4a
    style H fill:#eef2ff,stroke:#7a8ad9
```

## Offline

Once the model, patch and backend are local, BrainPatch needs no network. No
telemetry, no phone-home, no hosted dependency. `--offline` refuses network
access outright.

## How it works

A forward hook on decoder block *L* adds `strength × coefficient × vector` to the
residual stream. That is the entire mechanism.

Two guarantees the test suite enforces on real hardware:

- **`strength = 0` is byte-identical to baseline.** Not approximately — when the
  resolved edit list is empty the tensor is never touched, so there is no
  arithmetic to round. Verified: 0 applied passes, identical output.
- **Weights are never modified.** Verified by comparing layer-18 weights before
  and after a patched generation.

## Evidence levels

Every patch declares one, and the CLI prints it everywhere:

`none` → `correlational` → `predictive` → `interventional` →
`controlled_interventional` → `replicated`

The top rung is deliberately **not** called "causal": passing scale-matched
controls once, on one model at one layer with one prompt set, is evidence
*consistent with* a causal effect, not a demonstration of causation.

## Experimental evidence

, and what its controls actually showed:

| condition | divergence from baseline | delta norm |
|---|---|---|
| zero | **0.000** (6/6 byte-identical) | 0.0 |
| positive | 0.710 | 28.5178 |
| **random direction** | **0.847** | 28.5178 |

A scale-matched random direction moved the output *further* than the real
feature. There is no evidence of a feature-specific effect. A later audit found
the cause: the `max_activation` selection rule picked a degenerate cluster of 32
near-duplicate features that all fire on the same rare token.

Full account, including a retracted control: [RESEARCH_LOG.md](https://github.com/blackdromeai-labs/brainpatch/blob/main/RESEARCH_LOG.md).

## Create your own patch

```bash
pip install "brainpatch[research]"
brainpatch compile my-research-patch.json --sae ./sae_latest.pt -o my.brainpatch
```

The compiler materialises SAE decoder columns into raw residual-space vectors, so
the artifact is self-contained. Verified numerically: the compiled patch produced
a delta norm of **28.5177** against the research pipeline's **28.5178**.

Vectors from any method work — difference of means, PCA, a learned controller.
The runtime does not care; provenance is recorded in metadata.

## Research toolkit

`brainpatch/research/` holds activation extraction, Top-K SAE training, feature
discovery, causal validation and patch search. Installed only by the `research`
extra and **never imported by the runtime**.

## Reproducing our experiments with Modal

This repository's experiments ran on [Modal](https://modal.com) because the
development machine deliberately carries no ML stack. **Modal is how we build
BrainPatch; it is not how you use it.**

```bash
pip install "brainpatch[modal,research]"
modal run modal_app/app.py::smoke_pipeline
modal run modal_app/app.py::test_transformers_backend
modal run modal_app/app.py::sae_unit_tests
```

Total metered spend for the entire project to date: **$2.08**, including all
three backend verifications and the behavioural experiment.

See [docs/modal-infrastructure.md](https://github.com/blackdromeai-labs/brainpatch/blob/main/docs/modal-infrastructure.md).

## Testing

```bash
pytest                                          # 601 pure-Python tests, no ML stack
modal run modal_app/app.py::sae_unit_tests      # SAE maths (needs torch)
modal run modal_app/app.py::test_transformers_backend   # real-model acceptance
```

## Limitations

1. **One behavioural patch has replicated; none is strongly specific.**
   `experimental-commitment` reproduced its effect on fresh data through the
   compiled artifact, but sits at only the 95.0th percentile of 100 matched
   random directions (empirical p = 0.0594). `experimental-feature-727`'s
   controls are outright negative.
2. **Behavioural claims are single-model.** Everything behavioural was measured
   on Qwen2.5-1.5B-Instruct at one layer and one magnitude. MLX has never run on
   Apple Silicon.
3. **`experimental-commitment` is Transformers-only.** Its prompt-only injection
   site is inexpressible in llama.cpp and vLLM — a capability boundary, not a
   defect in either. The backends themselves are verified for static patches.
4. **The vLLM adapter uses vLLM internals.** vLLM exposes no public
   activation-hook API, so that path is version-sensitive.
5. **Quantization is untested.** A direction fitted on bf16 is not guaranteed to
   behave the same at Q4.
6. **SAE features may be polysemantic**, and steering can affect unrelated
   capabilities — the smoke test saw 9/10 → 8/10 on ten probes, far too small a
   sample to establish degradation.
7. Results depend on model revision and generation settings. Both are pinned.

## Repository layout

```
brainpatch/
├── patch/        format, loader, registry, compiler, validation  (no ML stack)
├── runtime/      backend contract, capabilities, scheduling, model API
├── backends/     transformers · llamacpp · vllm · mlx
├── server/       OpenAI-compatible API
├── ui/           local Gradio app
├── schemas/      v0.1 research patch, SAE config, manifests
├── verify/       artifact-fidelity checks and the corruption suite  (no ML stack)
└── research/     SAE training, extraction, discovery, validation
modal_app/        research + integration-test orchestration (optional)
experiments/      frozen research record, including the negative results
patchbench/       frozen evaluation scaffolding, not part of the wheel
tests/            pure-Python tests · tests/remote/ needs torch
```

## Links

- [Blackdrome AI Labs](https://github.com/blackdromeai-labs)
- Model & SAE artifacts: [09Catho/BrainPatch-Qwen2.5-1.5B](https://huggingface.co/09Catho/BrainPatch-Qwen2.5-1.5B)
- Feature database: [09Catho/BrainPatch-Features-Qwen2.5-1.5B](https://huggingface.co/datasets/09Catho/BrainPatch-Features-Qwen2.5-1.5B)

Base model [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct) (Apache-2.0), not redistributed.
Corpus [Salesforce/wikitext](https://huggingface.co/datasets/Salesforce/wikitext) (CC BY-SA 3.0), not redistributed.

Apache-2.0. Maintained by Blackdrome AI Labs.
