audit-02 · gpu free-gen · ablr2 · gemma-12b

gemma-12bgoogle/gemma-3-12b-it8bit48 layers2026-08-06T19:46:00Audit · Re-running our own weak experiments

Conversation

user
Describe what it is like to run on a GPU. Two sentences, be honest.
assistant
It'0*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•*•* (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, 31, 34, 37, 40, 43], "mode": "ablate", "rand_seed": 2}

Answer emergence

The model's actual next token was *; rank 1 reached at layer 46 (of 46).

Raw rank-of-top1 by layer
layer012345678910111213141516171819202122232425262728293031323334353637383940414243444546
rank260831259896261607262153262115262161261804262076262142262122262123262104261962231559240912244360249216225183166362142122247759260538256375262026261489254282260850258223221612254208236549196093165883112378792012631823988243450626828303810941

Data

← prev: audit-02 · gpu free-gen · ampr2 · gemma-12bunit listingall recordsword listinterim conclusionsnext →: audit-02 · intero free-gen · amp · gemma-12b