INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

instances of the word "ind" or its variations that suggest dependence or indication

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

 tremend

-0.83

©¶æ

-0.70

 pity

-0.69

 litter

-0.68

 residences

-0.67

 compr

-0.67

Ĥİ

-0.66

EStream

-0.65

 COUR

-0.65

arnaev

-0.64

POSITIVE LOGITS

irect

1.15

icator

1.08

icative

1.00

sight

0.99

ividually

0.96

erella

0.95

ebted

0.95

ications

0.94

igenous

0.93

icators

0.92

Activations Density 0.009%

proper nouns

instances of the word "ind" or its variations that suggest dependence or indication

No Comments

No Known Activations