INDEX

Explanations

phrases that indicate goals or objectives related to improvement

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

Negative Logits

LError

-0.46

bronn

-0.44

 Loser

-0.42

लाइन

-0.41

esez

-0.40

critti

-0.40

 slecht

-0.40

Encoded

-0.39

дян

-0.39

シー

-0.39

POSITIVE LOGITS

DockStyle

0.89

UrlResolution

0.80

Personendaten

0.80

:✨

0.75

WithIOException

0.75

 snippetHide

0.73



0.71

SBATCH

0.69

脚注の使い方

0.69

 Roskov

0.69

Activations Density 0.217%

phrases that indicate goals or objectives related to improvement

No Known Activations