INDEX

Explanations

comparative adjectives indicating a higher degree or level

oai_token-act-pair · gpt-3.5-turbo

phrases indicating an increase or a comparative magnitude

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

breaks

-0.91

Bio

-0.76

BO

-0.75

TED

-0.74

raid

-0.73

IB

-0.72

 Bleach

-0.72

SE

-0.71

pta

-0.71

PM

-0.70

POSITIVE LOGITS

 "$:/

1.00

 than

1.00

 eleph

0.96

 unbeliev

0.94

 proport

0.90

 abund

0.87

 magnification

0.87

 ende

0.85

 heights

0.85

 guiIcon

0.85

Activations Density 0.010%

comparative adjectives indicating a higher degree or level

phrases indicating an increase or a comparative magnitude

No Comments

No Known Activations