INDEX

Explanations

German language characters such as "Ã"

oai_token-act-pair · gpt-3.5-turbo

non-standard or corrupted characters in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

enegger

-0.90

 decomp

-0.70

iflower

-0.63

 blush

-0.62

 laugh

-0.61

 comprehens

-0.61

 flask

-0.61

 buckle

-0.60

kne

-0.60

schild

-0.59

POSITIVE LOGITS

ĺ

1.61

ģ

1.60

¢

1.59

ĵ

1.57

ļ

1.52

į

1.50

Ĩ

1.49

ħ

1.49

¤

1.49

³

1.48

Activations Density 0.015%

German language characters such as "Ã"

non-standard or corrupted characters in the text

No Comments

No Known Activations