INDEX

Explanations

numeric quantities

oai_token-act-pair · gpt-3.5-turbo

numerical indicators and measurements related to time and quantity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ĸļ

-1.03

 omission

-0.71

 Gillespie

-0.69

 shell

-0.69

 Redmond

-0.64

querque

-0.64

Bok

-0.64

hower

-0.63

 Moreno

-0.63

ĺ

-0.61

POSITIVE LOGITS

0.81

00000000

0.76

awatts

0.70

oldown

0.68

Sah

0.66

ocry

0.66

lde

0.65

annabin

0.65

 Spartans

0.64

Hz

0.63

Activations Density 0.231%

numeric quantities

numerical indicators and measurements related to time and quantity

No Comments

No Known Activations