INDEX

Explanations

mentions or references to hats

oai_token-act-pair · gpt-3.5-turbo

hyphenated adjectives and phrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Daylight

-0.83

 agile

-0.70

 Vaughn

-0.67

 Polaris

-0.67

 radial

-0.66

 subsistence

-0.66

 disposable

-0.66

DD

-0.66

 velocity

-0.66

 inertia

-0.65

POSITIVE LOGITS

filled

1.17

bub

1.14

taking

1.08

pee

1.07

dropping

1.07

seeking

1.07

drops

1.06

puff

1.05

cart

1.04

seekers

1.03

Activations Density 0.129%

mentions or references to hats

hyphenated adjectives and phrases

No Comments

No Known Activations