INDEX

Explanations

categorical labels and structured data formats

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

category: and source:

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

ftagPool

-0.81

 חיצוניים

-0.61

 zijne

-0.57

 Geſch

-0.56

нгред

-0.54

exitRule

-0.53

 بيها

-0.53

Vidite

-0.52

 dieſer

-0.52

 Winaray

-0.51

POSITIVE LOGITS

0.47

 Type

0.44

 type

0.40

 ISNI

0.40

KURZBESCHREIBUNG

0.38

ante

0.36

type

0.36

 Brandt

0.36

Vue

0.36

Activations Density 0.062%

categorical labels and structured data formats

category: and source:

No Comments

No Known Activations