INDEX

Explanations

phrases indicating the occurrence or location of events

oai_token-act-pair · gpt-3.5-turbo

instances of events or actions occurring

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

mini

-0.71

gian

-0.68

ãģķ

-0.65

Friend

-0.64

croft

-0.64

lite

-0.64

ipation

-0.63

tu

-0.63

 Refresh

-0.62

Haunted

-0.62

POSITIVE LOGITS

 place

2.04

place

1.38

 Place

1.24

 precedence

1.06

Place

1.03

 shape

1.03

 places

1.01

 effect

0.96

 bearings

0.86

 advantage

0.84

Activations Density 0.044%

phrases indicating the occurrence or location of events

instances of events or actions occurring

No Comments

No Known Activations