INDEX

Explanations

references to specific car brands

oai_token-act-pair · gpt-3.5-turbo

references to specific car brands and notable individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

bread

-0.85

ukemia

-0.85

 jars

-0.73

milo

-0.68

bats

-0.66

OGR

-0.66

sticks

-0.63

ous

-0.63

matically

-0.62

ies

-0.62

POSITIVE LOGITS

 Subaru

0.80

avorite

0.74

irit

0.74

prime

0.73

 Asuka

0.72

ministic

0.72

ropy

0.70

illes

0.69

0.68

ADRA

0.67

Activations Density 0.036%

references to specific car brands

references to specific car brands and notable individuals

No Comments

No Known Activations