INDEX

Explanations

perfor

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_11.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.11.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 invers

-0.28

rrha

-0.26

rin

-0.26

ray

-0.25

ROUT

-0.24

Arr

-0.24

 sank

-0.24

ãĥªãĥ³ãĤ°

-0.24

pen

-0.24

çĽ¸åıį

-0.24

POSITIVE LOGITS

STALL

0.32

è¦ģç´§

0.28

ãģĿãģĨãģł

0.28

SAFE

0.26

è¿ĩå¾Ĺ

0.26

 SAFE

0.25

 Palo

0.25

 Comfort

0.25

 Maul

0.25

ture

0.25

Activations Density 0.342%

perfor

No Comments

No Known Activations