INDEX

Explanations

brand names or proper nouns

oai_token-act-pair · gpt-3.5-turbo

empty or placeholder segments in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 inev

-0.68

matically

-0.67

voc

-0.66

 horr

-0.65

ajor

-0.63

aciously

-0.63

lessly

-0.63

ost

-0.62

izational

-0.62

 unres

-0.61

POSITIVE LOGITS

 Feet

0.95

 Ones

0.94

ness

0.93

idge

0.90

 Foods

0.89

 Guys

0.88

pedia

0.86

ly

0.86

 Hands

0.84

 Measures

0.84

Activations Density 0.152%

brand names or proper nouns

empty or placeholder segments in the text

No Comments

No Known Activations