INDEX

Explanations

personal pronouns and relational phrases indicating involvement or connection

oai_token-act-pair · gpt-3.5-turbo

expressions of personal feelings and opinions

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

 Bridgewater

-0.69

Amp

-0.65

 Toast

-0.64

 Verge

-0.64

 Redmond

-0.63

 Chao

-0.63

 Davies

-0.63

 Robinson

-0.62

 Eliot

-0.62

 Wellington

-0.61

POSITIVE LOGITS

âĢ

1.71

ï¸ı

1.18

Ò

1.13

Â

1.05

âĪ

1.05

â

1.00

âĶ

0.99

ðŁĺ

0.97

Ã±

0.97

Â»

0.96

Activations Density 0.283%

personal pronouns and relational phrases indicating involvement or connection

expressions of personal feelings and opinions

No Comments

No Known Activations