INDEX

Explanations

references to purchasing or upgrading a product or service

oai_token-act-pair · gpt-3.5-turbo

the word "Plus" indicating additional or supplementary information

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

adr

-1.07

ufact

-0.84

amental

-0.82

athed

-0.78

itory

-0.78

ural

-0.76

gments

-0.75

stem

-0.75

ruction

-0.75

istry

-0.73

POSITIVE LOGITS

Plus

0.89

henko

0.83

 Plus

0.81

++++

0.73

+++

0.70

ï¸

0.68

 PLUS

0.68

/-

0.68

 Bonus

0.65

atz

0.64

Activations Density 0.014%

references to purchasing or upgrading a product or service

the word "Plus" indicating additional or supplementary information

No Comments

No Known Activations