docs: add install and terminal chat quickstart

README.md

@@ -53,27 +53,78 @@ v2 differences from v1: the brevity gap on short prompts drops from 0.27 to
 - Holdout (4,968 rows): pearson_macro **0.9656**, mae 0.058.
 - Mapping battery (30 P1, 6 P2, 40 garbage): P1 gap 0.13, P2 direction 6/6.
 
-## Usage
+## Quickstart
+
+The output is a capability profile, not a distribution over models: each value
+is an independent sigmoid in `[0, 1]`.
+
+### Install
+
+```bash
+python3 -m venv .venv-systemone
+.venv-systemone/bin/python -m pip install --upgrade pip
+.venv-systemone/bin/python -m pip install --index-url https://download.pytorch.org/whl/cu128 \
+  torch==2.11.0
+.venv-systemone/bin/python -m pip install transformers==5.20.2 safetensors==0.8.0 \
+  tokenizers==0.23.3 numpy==1.26.4 huggingface-hub==2.2.0
+```
+
+Use `https://download.pytorch.org/whl/cu126` for driver 560-569, or
+`https://download.pytorch.org/whl/cpu` for CPU-only. The script below picks
+BF16 on CUDA and FP32 on CPU, as recommended.
+
+### Terminal chat
+
+Type a query, read the vector. Save as `chat.py` and run
+`.venv-systemone/bin/python chat.py`.
 
 ```python
 import torch
 from transformers import AutoModelForSequenceClassification, AutoTokenizer
 
-repo = "regolo/brick-sysone-V1"
-tok = AutoTokenizer.from_pretrained(repo)
+LABELS = ["instruction_following", "coding", "math_reasoning",
+          "world_knowledge", "planning_agentic", "creative_synthesis"]
+REVISION = "9263a397a366ee22921214ddeebab1f5e4e8facd"
+
+device = "cuda" if torch.cuda.is_available() else "cpu"
+dtype = torch.bfloat16 if device == "cuda" else torch.float32
 model = AutoModelForSequenceClassification.from_pretrained(
-    repo, dtype=torch.bfloat16
-).eval()
-
-batch = tok("debug this stacktrace...", return_tensors="pt", truncation=True, max_length=8192)
-with torch.inference_mode():
-    logits = model(**batch).logits
-print(torch.sigmoid(logits))
-# [instruction_following, coding, math_reasoning, world_knowledge, planning_agentic, creative_synthesis]
+    "regolo/brick-sysone-V1", revision=REVISION, dtype=dtype,
+    attn_implementation="sdpa",
+).to(device).eval()
+tokenizer = AutoTokenizer.from_pretrained("regolo/brick-sysone-V1", revision=REVISION)
+limit = min(model.config.max_position_embeddings, 8192)
+
+while True:
+    try:
+        query = input("query> ").strip()
+    except (EOFError, KeyboardInterrupt):
+        break
+    if not query:
+        continue
+    encoded = tokenizer(query, return_tensors="pt", truncation=True, max_length=limit).to(device)
+    with torch.inference_mode():
+        vector = model(**encoded).logits[0].sigmoid().float().tolist()
+    for label, value in zip(LABELS, vector):
+        print(f"  {label:<20} {value:.4f}")
+    print(f"  vector: [{', '.join(f'{v:.4f}' for v in vector)}]")
+```
+
+```
+query> write a python parser that streams csv rows
+  instruction_following 0.5117
+  coding                0.9258
+  math_reasoning        0.0903
+  world_knowledge       0.1157
+  planning_agentic      0.0630
+  creative_synthesis    0.0198
+  vector: [0.5117, 0.9258, 0.0903, 0.1157, 0.0630, 0.0198]
 ```
 
-Use FP32 on CPU and BF16 on CUDA. Truncate at 8192 tokens; longer inputs are
-outside the training distribution.
+Input is truncated at 8192 tokens; longer inputs are outside the training
+distribution. The Brick router classifies only the last user message, never
+system or assistant turns. Full scripts with `--json`, `--device` and one-shot
+mode: `examples/systemone/` in the Brick repository.
 
 ## License