INDEX

Explanations

terms related to wire or wiring

oai_token-act-pair · gpt-3.5-turbo

mentions of wire

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Modest

-0.70

ivia

-0.69

Italy

-0.65

 Serious

-0.63

 Answer

-0.63

Ca

-0.63

 Scand

-0.63

ãħĭ

-0.62

Specific

-0.62

Instance

-0.62

POSITIVE LOGITS

 wire

1.27

 wires

1.08

wire

1.01

 mesh

1.00

lessly

0.96

 wiring

0.95

kefeller

0.91

tap

0.86

oshenko

0.85

 Wire

0.85

Activations Density 0.005%

terms related to wire or wiring

mentions of wire

No Comments

No Known Activations