INDEX

Explanations

words related to famous towers or buildings

oai_token-act-pair · gpt-3.5-turbo

references to "Trump Tower."

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

é»Ĵ

-0.89

dit

-0.74

ãģ®éŃĶ

-0.72

milo

-0.71

Ð¼

-0.70

cause

-0.69

ãĤī

-0.67

ãĤ±

-0.67

ãĤŃ

-0.66

ktop

-0.63

POSITIVE LOGITS

 Towers

1.14

 skysc

0.98

 Tower

0.96

tower

0.94

 towers

0.89

fall

0.86

 tower

0.85

 Heist

0.82

 Tycoon

0.80

 balcony

0.78

Activations Density 0.010%

words related to famous towers or buildings

references to "Trump Tower."

No Comments

No Known Activations