INDEX

Explanations

words related to intentions or reasons, especially in the context of crime or conflict

oai_token-act-pair · gpt-3.5-turbo

references to motives behind actions or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ummer

-0.98

semble

-0.84

iannopoulos

-0.81

 Leopard

-0.77

thumbnails

-0.75

hem

-0.73

Lup

-0.72

scl

-0.72

rooms

-0.71

BuyableInstoreAndOnline

-0.70

POSITIVE LOGITS

 motives

1.04

 motive

1.04

 motivations

0.97

 motivation

0.92

 justifying

0.86

TextColor

0.86

 rationale

0.77

 reasoning

0.77

 motivated

0.73

why

0.73

Activations Density 0.020%

words related to intentions or reasons, especially in the context of crime or conflict

references to motives behind actions or events

No Comments

No Known Activations