INDEX

Explanations

mentions of religious figures, particularly priests

oai_token-act-pair · gpt-3.5-turbo

references to priests

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

DonaldTrump

-0.90

ilater

-0.82

issors

-0.70

BAT

-0.69

 Wonderland

-0.67

Ø§ÙĦ

-0.65

 Pebble

-0.65

RFC

-0.64

oday

-0.64

 disadvant

-0.64

POSITIVE LOGITS

esses

1.55

ess

1.15

 priest

1.10

 priests

1.09

 ordained

0.92

hood

0.81

 exorc

0.81

esan

0.80

ly

0.79

 Priest

0.78

Activations Density 0.025%

mentions of religious figures, particularly priests

references to priests

No Comments

No Known Activations