INDEX

Explanations

phrases indicating personal responsibility and ownership in experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

TintMode

-0.58

彼は

-0.56

ništvo

-0.56

彼女は

-0.51

 odkazy

-0.51

 Huguen

-0.48

phosa

-0.47

 Euphrates

-0.47

 они

-0.47

ronique

-0.46

POSITIVE LOGITS

 what

1.31

how

1.24

 whatever

1.24

 everything

1.16

whatever

1.04

everything

1.04

 AssemblyCulture

0.98

 where

0.98

why

0.97

 wherever

0.95

Activations Density 0.459%

phrases indicating personal responsibility and ownership in experiences

No Comments

No Known Activations