INDEX

Explanations

unintelligible characters or artifacts that might be related to an encoding or formatting issue

oai_token-act-pair · gpt-3.5-turbo

special characters or symbols in the document

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

 alleged

-0.86

 interrogation

-0.83

 stump

-0.83

 manif

-0.80

 disputed

-0.80

 habit

-0.79

 supposed

-0.79

 purported

-0.78

 marginal

-0.78

 unilateral

-0.78

POSITIVE LOGITS

ï¸ı

1.57

 Featuring

1.14

âĻ¥

1.10

Enjoy

1.07

ï¸

1.06

ðŁ

1.05

âľ

1.05

Created

1.03

Unique

1.01

Includes

1.01

Activations Density 0.452%

unintelligible characters or artifacts that might be related to an encoding or formatting issue

special characters or symbols in the document

No Comments

No Known Activations