INDEX

Explanations

references to people being taken somewhere or involving locations and actions taken towards them

oai_token-act-pair · gpt-3.5-turbo

instances of action or status updates regarding individuals or events

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

 incent

-0.79

 detecting

-0.76

 describ

-0.76

 frontline

-0.75

 speeding

-0.72

 reluct

-0.71

 prosec

-0.70

 honoured

-0.70

 quir

-0.69

 periodic

-0.68

POSITIVE LOGITS

Reviewer

1.31

.?

1.20

Loading

1.17

.-

1.15

âĸł

1.14

SOURCE

1.12

Article

1.11

[_

1.10

RAW

1.10

Advertisements

1.10

Activations Density 0.131%

references to people being taken somewhere or involving locations and actions taken towards them

instances of action or status updates regarding individuals or events

No Comments

No Known Activations