INDEX

Explanations

references to identity and questioning who someone is

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.12.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Architecture

jumprelu

Context Size

1,024

Dataset

chanind/pile-uncopyrighted-gemma-1024-abbrv-1B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

makeText

-0.68

StoryboardSegue

-0.66

 experiences

-0.60

exper

-0.60

 exper

-0.60

rishnan

-0.60

 Exper

-0.59

 menta

-0.58

 Penh

-0.58

<eos>

-0.58

POSITIVE LOGITS

 recevrez

0.91

 Normdatei

0.82

 Whence

0.79

antaranya

0.76

 hvem

0.75

 colectiva

0.72

 coute

0.70

 excepción

0.69

 Theſe

0.68

ndose

0.68

Activations Density 0.016%

references to identity and questioning who someone is

No Comments

No Known Activations

references to identity and questioning who someone is

No Comments

No Known Activations