INDEX

Explanations

references to conspiracy theories

oai_token-act-pair · gpt-3.5-turbo

references to conspiracy theories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rien

-0.83

Thom

-0.74

zl

-0.71

Environment

-0.69

endered

-0.68

then

-0.66

region

-0.65

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

-0.64

oven

-0.63

td

-0.63

POSITIVE LOGITS

 theories

1.10

 theorists

1.07

 conspiracy

1.07

 theorist

1.06

 conspir

0.98

 theory

0.87

 Conspiracy

0.86

 theor

0.86

eering

0.84

 creep

0.81

Activations Density 0.018%

references to conspiracy theories

references to conspiracy theories

No Comments

No Known Activations