INDEX

Explanations

references to religious or philosophical concepts and texts

oai_token-act-pair · gpt-3.5-turbo

references to religious or philosophical themes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Bloomberg

-0.89

tackle

-0.87

boarding

-0.85

-$

-0.81

CEO

-0.80

Todd

-0.79

Cash

-0.77

($

-0.77

Alert

-0.76

gc

-0.76

POSITIVE LOGITS

 latter

1.25

 slightest

1.24

 Greeks

1.12

 inhabitants

1.06

 greatest

1.06

 heavens

1.03

 earliest

1.00

ocratic

0.99

 twentieth

0.98

 remainder

0.98

Activations Density 0.915%

references to religious or philosophical concepts and texts

references to religious or philosophical themes

No Comments

No Known Activations