INDEX

Explanations

instructions or steps in a process

oai_token-act-pair · gpt-3.5-turbo

instructions or descriptors related to cooking processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 incentives

-0.85

 Kahn

-0.82

 successors

-0.80

 Promotion

-0.80

 Opportun

-0.77

 Achievements

-0.77

 Events

-0.74

 Messages

-0.74

 Achievement

-0.74

 Equity

-0.74

POSITIVE LOGITS

 moist

1.38

 thicker

1.33

 wrink

1.30

wet

1.29

 oily

1.24

dry

1.24

 swollen

1.21

 oxid

1.17

 melted

1.16

 coated

1.15

Activations Density 0.526%

instructions or steps in a process

instructions or descriptors related to cooking processes

No Comments

No Known Activations