INDEX

Explanations

the word 'have'

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate the occurrence or presence of events or conditions

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

cius

-0.62

buster

-0.61

immune

-0.59

icut

-0.58

 Niet

-0.58

illusion

-0.57

doing

-0.57

atron

-0.56

ï¸ı

-0.56

gypt

-0.56

POSITIVE LOGITS

aughtered

0.70

 plenty

0.67

everal

0.66

 occasions

0.64

enty

0.64

itia

0.64

ukong

0.63

 occas

0.62

iosyncr

0.62

ilege

0.62

Activations Density 0.040%

the word 'have'

phrases that indicate the occurrence or presence of events or conditions

No Comments

No Known Activations