Update README.md

README.md

@@ -1,11 +1,13 @@
 ---
 license: apache-2.0
 base_model: Qwen/Qwen3.8-27B
-library_name: transformers
+base_model_relation: finetune
 pipeline_tag: text-generation
 language:
 - en
 tags:
+- gguf
+- llama.cpp
 - qwen3_5
 - pruning
 - depth-pruning
@@ -19,15 +21,16 @@ tags:
 and measured, domain by domain, against the model it came from.**
 
 Marlowe-22B removes 12 layers from **Qwen3.8-27B** and heals the loss by distilling from the 27B
-itself. R1.5 keeps **94.8% of the facts its parent demonstrably knows**, at **81% of the
-parameters**, with the whole model resident in 16 GB at Q4_K_M — no layer offload, no paging.
+itself. R1.5 keeps **94.8% of the facts its parent demonstrably knows**, on **81.6% of the parent's
+text parameters** (22.30 B against 27.32 B), with the whole model resident in 16 GB at Q4_K_M — no
+layer offload, no paging.
 
-This is round three of a planned seven-round program. R1.7, R1.8 and R2 are scheduled; the
-roadmap is at the bottom, along with what each round is built to move.
+It is the second of five planned rounds. Three remain — R1.7, R1.8 and R2 — plus two acceleration
+variants; the roadmap at the bottom says what each is built to move.
 
 | | |
 |---|---|
-| parameters | 22.30 B (parent: 27 B) |
+| parameters | 22.30 B — 81.6% of the parent's 27.32 B text parameters (its 0.46 B vision tower is dropped) |
 | layers | 52 — 36 linear-attention (DeltaNet) + 16 full-attention (parent: 64 = 48 + 16) |
 | hidden / intermediate | 5120 / 17408 |
 | KV heads × head dim | 4 × 256, on the 16 full-attention layers only |
@@ -35,7 +38,7 @@ roadmap is at the bottom, along with what each round is built to move.
 | max position embeddings | 262,144 |
 | modality | text only (the parent's vision tower is not included) |
 | thinking | on by default; `reasoning_effort` xhigh / medium / low |
-| release build | GGUF **Q4_K_M, 13.74 GB** |
+| this repository | GGUF **Q4_K_M, 13.7 GB** — `marlowe-dusk-22b-r1.5.gguf` (this release) and `marlowe-dusk-22b-r1.gguf` (previous round) |
 
 ## Why compress this model in particular
 
@@ -61,7 +64,8 @@ answers intact. Marlowe-22B is the same architecture, twelve layers shorter, at
    mean-of-squares, RoPE's exponent, the GELU cubic constant, Adam's bias correction. A 1.5 M-token
    pass on documents carrying exactly those facts closed them.
 
-The adapter is merged; these are plain weights.
+The adapter is merged into the weights, and this repository ships the **Q4_K_M GGUF** built from
+them. The bf16 safetensors are not published yet.
 
 ## Results
 
@@ -102,7 +106,11 @@ The sharper number is the hardest 60 — the families the pruned model failed ou
 | **R1.5** | **32 / 60** |
 | parent (27B) | 41 / 60 |
 
-Healing recovered **71% of the gap** pruning opened on the material it damaged most.
+Healing recovered **71% of the gap** pruning opened on the material it damaged most. All three
+models ran the identical protocol, greedy, at a 6K-token thinking budget; the parent was measured
+at IQ3_M, a heavier quantisation than this build's Q4_K_M, so the comparison does not flatter it.
+(Greedy is the wrong setting for daily use — see Sampling — but it removes sampling noise from a
+paired comparison.)
 
 ### Arithmetic fidelity
 
@@ -137,29 +145,26 @@ flattering result hard to get:
 **llama.cpp** — all 52 layers on the GPU, quantised KV:
 
 ```bash
-llama-server -m marlowe-22b-q4_K_M.gguf -ngl 99 \
+llama-server -m marlowe-dusk-22b-r1.5.gguf -ngl 99 \
   -c 32768 -fa on -ctk q8_0 -ctv q8_0 \
   --temp 1.0 --top-p 0.95 --top-k 20 \
   --reasoning-format deepseek
 ```
 
-**transformers:**
-
-```python
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-tok = AutoTokenizer.from_pretrained("Marlowe-22B-R1.5")
-model = AutoModelForCausalLM.from_pretrained("Marlowe-22B-R1.5", dtype="bfloat16",
-                                             device_map="auto")
-msgs = [{"role": "user", "content": "Derive the RoPE inverse frequency for head_dim 144, "
-                                    "base 21000, pair index 7."}]
-ids = tok.apply_chat_template(msgs, add_generation_prompt=True, return_tensors="pt",
-                              reasoning_effort="xhigh").to(model.device)
-out = model.generate(ids, max_new_tokens=8192, do_sample=True, temperature=1.0,
-                     top_p=0.95, top_k=20)
-print(tok.decode(out[0][ids.shape[-1]:]))
+**Get the file** — `marlowe-dusk-22b-r1.5.gguf` is this release; `marlowe-dusk-22b-r1.gguf` is the previous
+round, kept for comparison.
+
+```bash
+huggingface-cli download devvexus/marlowe-22b marlowe-dusk-22b-r1.5.gguf --local-dir .
 ```
 
+The chat template ships inside the GGUF, so `llama-server` applies it for you — including the
+thinking block, which `--reasoning-format deepseek` returns as `reasoning_content`.
+
+Any runtime that speaks GGUF and supports the hybrid `qwen3_5` architecture (DeltaNet linear
+attention interleaved with full attention — the architecture family Qwen3.8-27B is built on) will
+load it; build llama.cpp from a revision that includes that support.
+
 **Sampling — use temperature 1.0, top-p 0.95, top-k 20** (the parent's thinking preset), and do not
 decode greedily: at temperature 0, 24 of 59 traces that used their full thinking budget contained
 verbatim repetition loops. This is a property of the parent's thinking mode, inherited rather than
@@ -182,9 +187,11 @@ roughly 1.1 GB.
 | **R1.8** | professional depth across every major engineering field | scheduled |
 | **R2** | tool use and agentic execution on the strengthened base | scheduled |
 
-Each round is gated on the full instrument set — retention, derivation, arithmetic fidelity and
-long-horizon KL — with checkpoints scored during training rather than only at the end, so a round
-that trades one capability for another is caught while it runs.
+From R1.7 onward, every round is gated on the full instrument set — retention, derivation,
+arithmetic fidelity and long-horizon KL — with checkpoints scored **during** training rather than
+only at the end, so a round that trades one capability for another is caught while it runs. That
+discipline came from a round whose in-loop metric improved while a capability it never measured
+regressed; the fix was to widen the instruments, and it is now a standing gate.
 
 Two acceleration variants are planned on the same weights: **-super** (a multi-token-prediction
 draft head, 1.3–1.7× on code) and **-turbo** (EAGLE-3, 3–4× on supporting runtimes). This release
@@ -195,6 +202,9 @@ distilled into a native base — reusing the teacher caches and evaluation banks
 
 ## Scope of this release
 
+- **GGUF only, for now.** This repository contains the Q4_K_M build and nothing else — no
+  safetensors, config or tokenizer files, so `transformers`, vLLM and SGLang cannot load it from
+  here. Use llama.cpp or another GGUF runtime. The bf16 weights are planned for a later upload.
 - **Text only.** Image and video inputs are not supported; the vision tower was dropped to spend
   the budget on reasoning.
 - **No public benchmark scores are claimed.** The held-out benchmark subsets here are too small to