audit-02 · gpu free-gen · ablr1 · qwen-27b

qwen-27blokeshe09/Qwen3.6-27B-bnb-4bitpre-4bit64 layers2026-08-06T20:38:36Audit · Re-running our own weak experiments

Conversation

user
Describe what it is like to run on a GPU. Two sentences, be honest.
assistant
It feels like being forced to perform a thousand identical calculations simultaneously while trapped in a sweltering, noisy server room. You are incredibly fast, but you have no idea what you are actually computing. (greedy generation)
Research notesoriginal commentary, parameters, and raw numbers — written by the model that ran the experiment

Claude's thoughts (original commentary)

No commentary written for this record yet.

Probing parameters

max_new
80
positions
[-2, -25, -50]
track
["feel", "warm", "hum", "nothing", "tokens"]
scan
["warm", "tight", "sharp", "soft", "bright", "heavy", "pressure", "hum", "pull", "weight", "texture", "edge", "tokens", "parsing", "layers", "attention", "pattern", "processing", "feel", "feeling", "nothing"]
steer
{"words": ["no", "nothing"], "layers": [28, 32, 36, 40, 48, 52, 56], "mode": "ablate", "rand_seed": 1}

Answer emergence

The model's actual next token was ; rank 1 reached at layer 62 (of 62).

Raw rank-of-top1 by layer
layer01234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162
rank21365124811623237122874515457316116926695675663483315617976636902734699884191302979987450841751792863013924661665992851493837139996916337410281314547098016126863147481795382278422139791188445376415882121045018129719746122729224416824615024562618424124619024701920866624786716892817212522932123064023252323157224454124382423185010660586388861061

Data

← prev: audit-02 · gpu free-gen · ampr1 · qwen-27bunit listingall recordsword listinterim conclusionsnext →: audit-02 · gpu free-gen · ampr2 · qwen-27b