INDEX

Explanations

phrases related to location or direction

oai_token-act-pair · gpt-3.5-turbo

prepositional phrases indicating locations or contexts

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

LEY

-0.68

avier

-0.65

ridor

-0.64

rar

-0.64

 Examiner

-0.63

ractor

-0.63

pathy

-0.63

ahime

-0.63

BILITIES

-0.62

POSITIVE LOGITS

 consecut

0.95

 respective

0.95

 respectively

0.87

 vying

0.86

 denominations

0.86

 varying

0.81

 populate

0.78

 warr

0.76

 imaginable

0.75

 subsequ

0.75

Activations Density 0.660%

phrases related to location or direction

prepositional phrases indicating locations or contexts

No Comments

No Known Activations