INDEX

Explanations

the word "a" at various intensities

oai_token-act-pair · gpt-3.5-turbo

repetitive use of the indefinite article "a."

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 documents

-0.68

Edit

-0.66

 applications

-0.63

 events

-0.63

IDs

-0.62

 assemblies

-0.62

 articles

-0.62

 associations

-0.62

 Enter

-0.61

 raids

-0.61

POSITIVE LOGITS

tad

1.29

bit

1.18

kward

1.10

versive

1.04

urally

1.04

lot

1.00

rouse

0.99

verse

0.98

ided

0.98

jar

0.96

Activations Density 0.225%

the word "a" at various intensities

repetitive use of the indefinite article "a."

No Comments

No Known Activations