INDEX

Explanations

phrases related to institutions or specific entities named "Case"

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "Case"

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

pez

-1.04

orough

-0.95

¥ŀ

-0.95

resent

-0.88

 destro

-0.86

irted

-0.83

edged

-0.82

ĻĤ

-0.80

merce

-0.80

 forbidden

-0.80

POSITIVE LOGITS

 Case

1.45

cases

1.30

Case

1.18

case

1.12

 Cases

0.99

 CASE

0.97

 Files

0.86

 Keen

0.85

 Codes

0.80

 Carrier

0.80

Activations Density 0.009%

phrases related to institutions or specific entities named "Case"

mentions of the word "Case"

No Comments

No Known Activations