INDEX

Explanations

phrases involving pronouns and direct speech

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate direct addresses or commands

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Âł Âł Âł Âł

-0.71

 Orange

-0.65

ĵĺ

-0.62

 ï¿½

-0.62

 appellant

-0.60

ä½¿

-0.59

 pione

-0.58

DAY

-0.57

èĢħ

-0.57

ç«

-0.57

POSITIVE LOGITS

're

1.47

'll

1.42

've

1.32

'd

1.24

'm

1.20

't

1.04

's

0.95

Enlarge

0.88

sha

0.81

',"

0.81

Activations Density 0.385%

phrases involving pronouns and direct speech

phrases that indicate direct addresses or commands

No Comments

No Known Activations