INDEX

Explanations

instances of the word "of" followed by numbers

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "of."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

wcs

-0.79

Ô

-0.69

 Cosponsors

-0.68

 teasp

-0.67

ãĤ¼ãĤ¦ãĤ¹

-0.64

Fax

-0.64

phasis

-0.64

few

-0.63

ucci

-0.62

Password

-0.61

POSITIVE LOGITS

 varying

0.66

 differing

0.65

 ours

0.62

icial

0.62

 assorted

0.61

 disparate

0.61

 themselves

0.60

 souls

0.59

 course

0.59

 kindred

0.58

Activations Density 0.458%

instances of the word "of" followed by numbers

occurrences of the word "of."

No Comments

No Known Activations