INDEX

Explanations

logging-related terms and error messages from a system

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

Efq

-0.86

 myſelf

-0.82

</tfoot>

-0.79

ThemeOverlay

-0.79

ſelves

-0.78

 itſelf

-0.76

 Majefty

-0.76

ControllerAdvice

-0.76

StructEnd

-0.75

writeFieldEnd

-0.73

POSITIVE LOGITS

PRNewswire

0.48

 vitro

0.47

 impré

0.46

 tand

0.45

artal

0.43

say

0.43

 verso

0.43

jmniej

0.42

vis

0.42

 compromisos

0.41

Activations Density 0.012%

logging-related terms and error messages from a system

No Comments

No Known Activations