Instructions to use infosave/Ternary-Bonsai-2-27B-cmf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- cortiq
How to use infosave/Ternary-Bonsai-2-27B-cmf with cortiq:
# one Rust binary, no additional dependencies cargo install cortiq-cli # or a prebuilt binary from github.com/infosave2007/cmf/releases hf download infosave/Ternary-Bonsai-2-27B-cmf --include "*.cmf" --local-dir . ls *.cmf # some repos ship more than one quantization
cortiq run FILE.cmf --prompt "What is the capital of France?"
cortiq serve FILE.cmf --port 8080 # OpenAI-compatible server
- Notebooks
- Google Colab
- Kaggle
|
Download evaluation/SCORE_RECIPE.md from infosave/Ternary-Bonsai-2-27B-cmf: direct link, hf CLI and curl.
- Browser
- Download file 2.02 kB
-
https://huggingface.co/infosave/Ternary-Bonsai-2-27B-cmf/resolve/main/evaluation/SCORE_RECIPE.md
- Command line
-
hf download hf://infosave/Ternary-Bonsai-2-27B-cmf/evaluation/SCORE_RECIPE.md
-
curl -L -o SCORE_RECIPE.md https://huggingface.co/infosave/Ternary-Bonsai-2-27B-cmf/resolve/main/evaluation/SCORE_RECIPE.md
2.02 kB
| # Frozen evaluation recipe | |
| The fixtures in this directory are run-authored engineering diagnostics, | |
| released under the CMF Apache-2.0 terms. They are not a general benchmark. | |
| `prompts.jsonl` contains the eight frozen text prompts. The likelihood corpus | |
| and its token IDs are separate fixtures; do not retokenize with a different | |
| model or add a BOS token. | |
| ## Official reference | |
| The official PQ2 reference used llama.cpp/Prism release commit | |
| `9a9394a895b96003ca842a6041cb28ac49a108f7` with CUDA 12.8. The small | |
| `reference-teacher-score.cpp` helper links against the existing official | |
| `libllama`/ggml shared libraries and uses `llama_decode` plus | |
| `llama_get_logits_ith`; it does not build llama.cpp or redistribute its | |
| headers/libraries. Build the helper against that release's headers and | |
| libraries, then run: | |
| ```sh | |
| ./llama-teacher-score \ | |
| /path/to/Ternary-Bonsai-2-27B-PQ2_0.gguf \ | |
| evaluation/quality-corpus-token-ids.txt \ | |
| evaluation/reference-teacher-score.jsonl | |
| ``` | |
| The contract is two independent 512-input-token resets at starts 0 and 512, | |
| scoring the actual next IDs at targets 1..1024, with no synthetic BOS. The | |
| shipped raw JSONL records all 1,024 correct-token log-probabilities and top-1 | |
| IDs without a full-vocabulary dump. Reference results: mean NLL | |
| `1.81274952826`, PPL `6.12727139849`, top-1 `603/1024`. | |
| ## CMF affine comparison | |
| Build the patched runtime and its `score_fixed` example using | |
| `source/SOURCE_BUILD.md`, then use the same token-ID file and the model file: | |
| ```sh | |
| export XDG_RUNTIME_DIR=/tmp WGPU_BACKEND=vulkan CMF_GPU=wgpu | |
| export CMF_Q2TP_GPU=1 CMF_GPU_PROBE=0 CMF_MTP=0 | |
| ./target/debug/examples/score_fixed \ | |
| Ternary-Bonsai-2-27B-Q2TP-Affine.cmf \ | |
| evaluation/quality-corpus-token-ids.txt | |
| ``` | |
| The validated affine Vulkan output was total NLL `1856.528128685`, mean NLL | |
| `1.813015751`, PPL `6.128902833`; CPU mean NLL was `1.812992284`. Ordinary | |
| Q2TP is retained only as a comparison: mean NLL `1.886951899`, delta | |
| `+0.074202371`, outside the adopted `+0.05` primary gate. | |