INDEX

Explanations

phrases indicating a repeated action or event

oai_token-act-pair · gpt-3.5-turbo

phrases emphasizing repetition or recurrence

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

VICE

-0.69

Thirty

-0.65

antz

-0.65

 Theft

-0.63

asty

-0.63

 Serious

-0.61

tein

-0.59

 Compatibility

-0.59

Unix

-0.58

LIN

-0.58

POSITIVE LOGITS

 again

1.16

AGA

0.98

etheless

0.95

again

0.88

 until

0.83

 Again

0.76

 thereafter

0.75

 till

0.72

until

0.70

 anew

0.68

Activations Density 0.060%

phrases indicating a repeated action or event

phrases emphasizing repetition or recurrence

No Comments

No Known Activations