INDEX

Explanations

concepts related to fear

oai_token-act-pair · gpt-3.5-turbo

references to fear and anxiety in various contexts

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

•layer 11 - fear

No Comments

Negative Logits

nice

-0.77

odore

-0.65

arbon

-0.65

 authenticated

-0.64

 Prospect

-0.63

 contrace

-0.60

 Cowboy

-0.59

Vie

-0.59

 Contract

-0.58

 concess

-0.57

POSITIVE LOGITS

mong

1.49

lessness

1.32

lessly

1.26

fulness

1.05

crow

1.00

 Mong

1.00

fully

0.97

warts

0.92

 lest

0.88

wart

0.86

Activations Density 0.050%

concepts related to fear

references to fear and anxiety in various contexts

No Comments

No Known Activations