INDEX

Explanations

mentions of information or facts

oai_token-act-pair · gpt-3.5-turbo

phrases concerning common knowledge or widely accepted beliefs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

oubted

-0.93

isance

-0.90

phasis

-0.87

orthy

-0.86

oufl

-0.84

elled

-0.84

 teasp

-0.83

vous

-0.82

ealous

-0.81

arsh

-0.81

POSITIVE LOGITS

 Donald

1.08

 Barack

1.03

 Elon

1.03

 Kanye

0.97

 Hillary

0.95

 Bernie

0.94

 NASCAR

0.92

 LeBron

0.90

 Mitt

0.90

 Conor

0.89

Activations Density 0.583%

mentions of information or facts

phrases concerning common knowledge or widely accepted beliefs

No Comments

No Known Activations