INDEX

Explanations

references to something that will happen later

oai_token-act-pair · gemini-2.0-flash

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_18/width_16k/average_l0_13

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

TagMode

-0.69

tagHelper

-0.65

setVerticalGroup

-0.64

 שוליים

-0.59

 مشين

-0.58

SourceChecksum

-0.55

gants

-0.53

RenderAtEndOf

-0.52

ToBounds

-0.52

posables

-0.52

POSITIVE LOGITS

 previous

1.84

previous

1.70

 Previous

1.48

Previous

1.47

 earlier

1.43

 précédente

1.41

 précédentes

1.34

 precedenti

1.33

 precedente

1.32

 précédent

1.31

Activations Density 2.284%

references to something that will happen later

previous

No Comments

No Known Activations