INDEX

Explanations

adjectives describing behaviors or characteristics of people

oai_token-act-pair · gpt-3.5-turbo

phrases indicating common behaviors or characteristics of people

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

bernatorial

-0.79

=[

-0.76

ospons

-0.75

everal

-0.69

 Login

-0.69

 Anniversary

-0.68

 Scrib

-0.68

stantial

-0.66

 Pastebin

-0.65

nesday

-0.65

POSITIVE LOGITS

 afraid

1.25

 tempted

1.13

 obsessed

1.12

 unwilling

1.10

 reluctant

1.09

 fools

1.08

 unaware

1.07

 fooled

1.07

 fascinated

1.06

 intimidated

1.06

Activations Density 0.238%

adjectives describing behaviors or characteristics of people

phrases indicating common behaviors or characteristics of people

No Comments

No Known Activations