GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
agu18dec
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
agu18dec
Model Tree
Qwen/Qwen3.6-27B
Input Modalities
Output Modalities
Supported Functionality
| knob | value |
|---|---|
| base model | Qwen/Qwen3.6-27B (text decoder) |
| injection site | the embedding slot of marker token ㈜ (id 158983), flanked by <activation>/</activation> (neighbor ids 29, 510) |
| injection | replace the marker's embedding with 16000 · h / ‖h‖ (unit transform, α=16000; scale-free) |
| supported layers | residual stream layers 20–60, step 4 (11 layers) |
| prompt | the checkpoint's own concepts_raw template (in nla_meta.yaml): "An activation vector from layer 20 of a language model is enclosed in activation tags: <activation>㈜</activation>. Produce distinct concepts that encode this activation, each as a '- ' bullet on its own line." rendered with the chat template, add_generation_prompt=True |
| parse | one concept per - bullet line |
| eval sampling used in our benchmarks | k=1, T=1.0, top-p 0.95, top-k 64, max-new 256 |
python
import torchfrom transformers import AutoModelForCausalLM, AutoTokenizerfrom peft import PeftModelbase = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3.6-27B", torch_dtype=torch.bfloat16, device_map="cuda")tok = AutoTokenizer.from_pretrained("Qwen/Qwen3.6-27B")model = PeftModel.from_pretrained(base, "<this repo>")TEMPLATE = ("An activation vector from layer 20 of a language model is enclosed in activation ""tags: <activation>㈜</activation>. Produce distinct concepts that encode this ""activation, each as a '- ' bullet on its own line.")ids = tok.apply_chat_template([{"role": "user", "content": TEMPLATE}],tokenize=True, add_generation_prompt=True)mp = ids.index(158983) # the marker sloth = ... # your residual vector, layer in {20,24,...,60}v = 16000.0 * h / h.norm()embed = model.get_input_embeddings()inputs_embeds = embed(torch.tensor([ids], device="cuda"))inputs_embeds[0, mp] = v.to(inputs_embeds.dtype).to("cuda") # REPLACE the slot embeddingout = model.generate(inputs_embeds=inputs_embeds, do_sample=True, temperature=1.0,top_p=0.95, top_k=64, max_new_tokens=256)print(tok.decode(out[0], skip_special_tokens=True))
Judged baseline macro 0.785 over the 12 bank families (association 0.88, multihop 0.89, multihop-mt 0.87, multilingual 1.00, typo 1.00); hillclimbing: conjunctive 0.60, role-bound association 0.85, moral-committed 0.81, moral-deliberative 0.62, user-modeling 0.44 (k=1). Notable qualitative reads: never-written arithmetic intermediates, both poles of moral dilemmas from the empty assistant prefix, doctrine retrieval (Tarasoff, Double Effect) at layer 20, and latent user-attribute inference (see the workspace-bench records).
ckpts/ao/distill/final.s0/step105.adapter_config.json, adapter_model.safetensors (PEFT LoRA), meta.json
(training summary), nla_meta.yaml (the prompt/injection sidecar — the source of truth
for the load contract), run_config.yaml (full trainer config).agu18dec/local-workspace → ckpts/ao/rl/iolens.final.ddp600.s0/.