INDEX

Explanations

financial or numeric information in text data

oai_token-act-pair · gpt-3.5-turbo

instances of the word "total" in various contexts

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

ĪĴ

-0.75

maid

-0.74

heid

-0.71

 Briggs

-0.71

©¶æ

-0.68

yer

-0.67

paces

-0.66

bley

-0.66

selves

-0.66

rir

-0.64

POSITIVE LOGITS

itarian

1.16

 amount

0.92

itar

0.89

 length

0.80

catentry

0.79

 eclipse

0.79

 Eclipse

0.77

 number

0.75

 compensation

0.75

 quantity

0.71

Activations Density 0.033%

financial or numeric information in text data

instances of the word "total" in various contexts

No Comments

No Known Activations