INDEX

Explanations

information related to aides assisting individuals in various positions of power

oai_token-act-pair · gpt-3.5-turbo

mentions of aides or assistants in various contexts

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

aval

-0.73

itiveness

-0.70

urrency

-0.68

Trend

-0.64

rights

-0.63

PK

-0.63

odus

-0.63

origin

-0.61

eros

-0.61

ALE

-0.60

POSITIVE LOGITS

 aide

1.19

glers

1.05

 aides

1.01

maid

1.00

 Huma

0.99

 counsel

0.83

 staffer

0.81

eteria

0.80

wife

0.72

 adviser

0.71

Activations Density 0.005%

information related to aides assisting individuals in various positions of power

mentions of aides or assistants in various contexts

No Comments

No Known Activations