INDEX

Explanations

words related to insulating or emulating

oai_token-act-pair · gpt-3.5-turbo

words related to regulation and guidelines

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

 dimensional

-0.71

Es

-0.69

Veter

-0.66

Corp

-0.61

atom

-0.60

 bitters

-0.58

 meteor

-0.58

ne

-0.57

zan

-0.57

 plutonium

-0.57

POSITIVE LOGITS

ulate

1.74

ulating

1.57

ulates

1.52

ulated

1.36

ulation

1.34

ulators

1.18

ulations

1.17

ulator

1.17

ulatory

1.08

ulative

0.94

Activations Density 0.022%

words related to insulating or emulating

words related to regulation and guidelines

No Comments

No Known Activations