INDEX

Explanations

commands or prompts instructing the reader to take an action

oai_token-act-pair · gpt-3.5-turbo

references to personal ownership or possession

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

apo

-0.89

forth

-0.76

 Goes

-0.72

 Cohn

-0.70

/-

-0.66

avis

-0.65

ylum

-0.64

Pac

-0.63

ĸļ

-0.62

ilk

-0.62

POSITIVE LOGITS

own

1.31

 favourite

1.12

 favorite

1.01

ocard

0.91

selves

0.91

OWN

0.90

anmar

0.90

 imagination

0.89

self

0.89

 browser

0.81

Activations Density 0.105%

commands or prompts instructing the reader to take an action

references to personal ownership or possession

No Comments

No Known Activations