INDEX

Explanations

adjectives and nouns related to characteristics and attributes

oai_token-act-pair · gpt-3.5-turbo

concepts related to societal flaws and moral dilemmas

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

OWS

-0.68

ares

-0.64

 Picks

-0.64

ãĤ´ãĥ³

-0.62

Ĥİ

-0.61

catentry

-0.61

 Dreams

-0.59

 ç¥ŀ

-0.59

_-

-0.58

would

-0.58

POSITIVE LOGITS

 lurking

1.14

 involved

0.99

 waiting

0.98

 underway

0.95

 attached

0.92

 available

0.92

 awaiting

0.85

 underneath

0.84

 behind

0.84

 happening

0.83

Activations Density 0.440%

adjectives and nouns related to characteristics and attributes

concepts related to societal flaws and moral dilemmas

No Comments

No Known Activations