INDEX

Explanations

differences in versions or updates

oai_token-act-pair · gpt-3.5-turbo

repeated instances of a specific symbol or character sequence

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

 dirt

-0.76

ierrez

-0.70

 torn

-0.69

 rooting

-0.69

ggies

-0.68

 planted

-0.67

gad

-0.65

 scattering

-0.65

anwhile

-0.65

 cane

-0.64

POSITIVE LOGITS

¹

1.03

º

1.01

£

1.00

catentry

0.96

¬

0.95

¡

0.95

į

0.95

ı

0.92

»

0.89

Ķ

0.89

Activations Density 0.610%

differences in versions or updates

repeated instances of a specific symbol or character sequence

No Comments

No Known Activations