INDEX

Explanations

instances of punctuation or symbols, particularly those indicating a pause or list in writing

oai_token-act-pair · gpt-4o-mini Triggered by @bot

comparisons and contrasts

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 respectively

-0.39

Sef

-0.39

之

-0.38

 followed

-0.37

 SAFE

-0.36

ների

-0.36

Safe

-0.35

BASELINE

-0.35

 pena

-0.35

 Safe

-0.34

POSITIVE LOGITS

 otomatig

0.71

aarrggbb

0.66

 ujednoznacz

0.62

EndContext

0.57

bootstrapcdn

0.54

GEBURTSDATUM

0.54

 utafitiHapana

0.54

 mijne

0.51

Ours

0.49

AsUp

0.48

Activations Density 0.024%

instances of punctuation or symbols, particularly those indicating a pause or list in writing

comparisons and contrasts

No Comments

No Known Activations