INDEX

Explanations

words ending in 's

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter 's'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

mass

-0.68

EStream

-0.67

 saline

-0.63

Kiw

-0.62

Pwr

-0.62

bush

-0.62

Picture

-0.62

pour

-0.60

PUT

-0.59

 Slate

-0.59

POSITIVE LOGITS

pecially

1.02

selves

1.02

outhern

0.95

own

0.92

ELF

0.91

ources

0.87

ledged

0.85

outheast

0.83

atisf

0.83

ullivan

0.82

Activations Density 0.198%

words ending in 's

occurrences of the letter 's'

No Comments

No Known Activations