INDEX

Explanations

specific entities or brands followed by the possessive form and a number

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter 's' in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Lauder

-0.75

endez

-0.69

Compare

-0.68

lehem

-0.67

uate

-0.67

rette

-0.65

roup

-0.62

ovo

-0.62

ate

-0.62

 Compare

-0.61

POSITIVE LOGITS

own

1.20

ELF

1.12

 reputation

1.00

 shortcomings

0.96

 newest

0.95

 existence

0.94

 involvement

0.93

 demise

0.92

 failings

0.88

 intentions

0.88

Activations Density 0.206%

specific entities or brands followed by the possessive form and a number

occurrences of the letter 's' in various contexts

No Comments

No Known Activations