INDEX

Explanations

negative expressions or sentiments in the text

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

<eos>

-0.59

on

-0.56

pen

-0.50

-0.49

-0.48

ablon

-0.48

 Martinez

-0.48

oxin

-0.47

No

-0.47

POSITIVE LOGITS

ſelf

1.18

 Monfieur

1.13

 auffi

1.09

ſelves

1.08

 itſelf

1.08

 myſelf

1.08

 houſe

1.05

ſy

1.05

 iſt

1.04

 ſche

1.02

Activations Density 0.004%

negative expressions or sentiments in the text

No Known Activations