INDEX

Explanations

names of individuals, particularly those associated with notable achievements or roles

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

 pleaſure

-0.96

 purpoſe

-0.92

 poffible

-0.91

 Monfieur

-0.91

 itſelf

-0.90

 ་་

-0.90

 uſed

-0.89

 houſe

-0.88

 Jefus

-0.88

 myſelf

-0.88

POSITIVE LOGITS

Bo

0.60

ra

0.58

di

0.57

Sa

0.57

len

0.57

ap

0.57

Par

0.57

Pa

0.55

ma

0.54

Jo

0.54

Activations Density 0.135%

names of individuals, particularly those associated with notable achievements or roles

No Comments

No Known Activations