INDEX

Explanations

years denoted between 2011 and 2013

oai_token-act-pair · gpt-3.5-turbo

occurrences of the year 2013 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 @ 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

flo

-0.89

sembly

-0.75

warts

-0.72

merce

-0.68

srf

-0.67

 subsequ

-0.66

iaz

-0.65

bip

-0.64

 phosphate

-0.63

Commerce

-0.63

POSITIVE LOGITS

rd

0.85

-'

0.84

bard

0.83

å¹

0.83

 onwards

0.80

externalActionCode

0.69

etheus

0.69

 INTER

0.68

 onward

0.67

ennial

0.66

Activations Density 0.024%

years denoted between 2011 and 2013

occurrences of the year 2013 in various contexts

No Comments

No Known Activations