INDEX

Explanations

phrases related to a specific organization or brand

oai_token-act-pair · gpt-3.5-turbo

references to specific organizations or entities, particularly related to academia or research

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

 panc

-0.84

 Hillary

-0.72

 soup

-0.72

 Columbus

-0.68

 Chop

-0.68

Bos

-0.67

 Dunk

-0.65

 Vers

-0.65

 Tunis

-0.63

 boiled

-0.63

POSITIVE LOGITS

RL

4.68

rl

1.63

RL

1.62

RS

1.44

JR

1.33

LR

1.31

RC

1.30

SL

1.25

LL

1.22

ARA

1.18

Activations Density 0.010%

phrases related to a specific organization or brand

references to specific organizations or entities, particularly related to academia or research

No Comments

No Known Activations