INDEX

Explanations

proper names

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Harrison."

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

ãĥĨ

-0.75

 defaults

-0.72

*/

-0.70

*/

-0.68

Å

-0.68

 python

-0.68

opt

-0.67

ãĥİ

-0.67

ABS

-0.66

mag

-0.66

POSITIVE LOGITS

 Harrison

3.85

arrison

2.10

 Harding

1.60

 Garrison

1.29

 Frazier

1.25

 Hampton

1.24

 Hayward

1.23

 Hancock

1.19

 Fraz

1.19

 Reeves

1.19

Activations Density 0.015%

proper names

mentions of the name "Harrison."

No Comments

No Known Activations