INDEX

Explanations

phrases indicating relationships or connections

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

 fhew

-0.98

 Majefty

-0.93

 pleaſure

-0.91

 myſelf

-0.90

ftance

-0.86

 ſtate

-0.85

 ſeveral

-0.85

 uſed

-0.84

 themſelves

-0.81

 Diſ

-0.81

POSITIVE LOGITS

parsedMessage

0.72

PreferredItem

0.68

 CanadaChoose

0.61

</b>

0.57

Tos

0.57

reportWebVitals

0.57

MigrationBuilder

0.56

urlopen

0.56

IRUS

0.56

Tembelea

0.53

Activations Density 0.125%

phrases indicating relationships or connections

No Comments

No Known Activations