INDEX

Explanations

phrases related to legal terminology and evidential claims

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

 berlebihan

-0.52

 nicio

-0.52

欖

-0.49

 grecque

-0.49

emploi

-0.48

koľvek

-0.48

发表于

-0.48

 cotone

-0.48

ľ

-0.48

 ocasiones

-0.47

POSITIVE LOGITS

 للمعارف

0.68

EndProject

0.65

 Ouverture

0.65

 cherchés

0.64

 виправивши

0.62

RUnlock

0.61

Carcinogenicity

0.60

.")]

0.59

 linkovi

0.57

strando

0.56

Activations Density 0.096%

phrases related to legal terminology and evidential claims

No Comments

No Known Activations

phrases related to legal terminology and evidential claims

No Comments

No Known Activations