INDEX

Explanations

proper nouns or aliases

oai_token-act-pair · gpt-3.5-turbo

references to notable individuals or celebrities

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

Ö¼

-0.92

Īè

-0.76

lines

-0.73

matically

-0.71

ĻĤ

-0.68

rals

-0.68

 utilities

-0.67

relations

-0.65

wolves

-0.65

encers

-0.64

POSITIVE LOGITS

ichi

1.11

unta

0.95

 Haram

0.87

Ã±a

0.84

Ã±

0.83

aka

0.83

á¹

0.82

oka

0.82

ishi

0.80

ÅŁ

0.79

Activations Density 0.019%

proper nouns or aliases

references to notable individuals or celebrities

No Comments

No Known Activations