INDEX

Explanations

prominent words or phrases in image captions

oai_token-act-pair · gpt-3.5-turbo

instructions or commands related to multimedia elements

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rals

-0.61

said

-0.60

Rai

-0.60

comes

-0.59

meal

-0.58

lets

-0.57

fulness

-0.57

endment

-0.57

warm

-0.55

naires

-0.55

POSITIVE LOGITS

 Image

0.97

 Photo

0.76

UTERS

0.75

NPR

0.69

ertodd

0.66

 image

0.65

 toggle

0.65

iframe

0.64

¥ŀ

0.64

 PHOTO

0.63

Activations Density 0.019%

prominent words or phrases in image captions

instructions or commands related to multimedia elements

No Comments

No Known Activations