INDEX

Explanations

discussions or mentions of arrests in a text

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "arrest" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

erker

-0.65

yg

-0.64

este

-0.62

ocol

-0.62

wired

-0.62

onen

-0.60

box

-0.60

gha

-0.59

culus

-0.59

mor

-0.59

POSITIVE LOGITS

 arrests

3.85

 arrest

2.52

 Arrest

2.14

 prosecutions

1.99

 arrested

1.80

 convictions

1.69

 suspensions

1.63

 arresting

1.60

 raids

1.58

 seizures

1.55

Activations Density 0.016%

discussions or mentions of arrests in a text

occurrences of the word "arrest" and its variations

No Comments

No Known Activations