INDEX

Explanations

concepts related to distrust and mistrust in various contexts

oai_token-act-pair · gpt-3.5-turbo

terms related to distrust and skepticism towards authority and institutions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ammy

-0.83

Redditor

-0.83

gran

-0.81

education

-0.68

âĹ¼

-0.67

Dispatch

-0.65

acca

-0.65

ODE

-0.64

erry

-0.64

cise

-0.64

POSITIVE LOGITS

lessly

1.20

fully

1.11

ful

1.01

fulness

0.94

ingly

0.85

lessness

0.83

worthiness

0.82

uous

0.82

uously

0.80

yip

0.79

Activations Density 0.012%

concepts related to distrust and mistrust in various contexts

terms related to distrust and skepticism towards authority and institutions

No Comments

No Known Activations