INDEX

Explanations

keywords related to mistakes and poor decisions

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

ValueStyle

-0.59

DIPSETTING

-0.56

tal

-0.52

WebElementEntity

-0.52

종

-0.49

Tracce

-0.49

 doPost

-0.48

凌

-0.47

centers

-0.47

Concer

-0.47

POSITIVE LOGITS

 unwise

0.76

 imprud

0.71

BeginContext

0.68

 blunder

0.66

 foolish

0.64

lück

0.64

tagHelper

0.61

 mistakes

0.60

 stupidly

0.59

 risky

0.59

Activations Density 0.127%

keywords related to mistakes and poor decisions

No Comments

No Known Activations