INDEX

Explanations

phrases indicating a particular focus or emphasis within a statement or argument

oai_token-act-pair · gpt-3.5-turbo

references to specific situations or contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

inx

-0.79

atin

-0.68

ulner

-0.67

acers

-0.67

ricanes

-0.64

icons

-0.64

stars

-0.62

assets

-0.61

effects

-0.60

tested

-0.60

POSITIVE LOGITS

 regard

1.78

 respect

1.44

 vein

1.43

 case

1.38

 instance

1.37

 manner

1.32

 circumstance

1.28

 context

1.26

 situation

1.20

 regards

1.19

Activations Density 0.080%

phrases indicating a particular focus or emphasis within a statement or argument

references to specific situations or contexts

No Comments

No Known Activations