INDEX

Explanations

phrases related to programming or coding

oai_token-act-pair · gpt-3.5-turbo

prepositions and phrases indicating relationships or connections

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

 Sorceress

-0.71

arthed

-0.67

sson

-0.66

 Nadu

-0.63

Oracle

-0.63

 Mages

-0.63

hler

-0.61

NING

-0.61

riors

-0.60

blance

-0.60

POSITIVE LOGITS

pless

0.73

ordinate

0.66

=-=-

0.64

anthrop

0.62

animate

0.61

enes

0.60

Zip

0.60

brid

0.60

anz

0.59

direct

0.58

Activations Density 0.651%

phrases related to programming or coding

prepositions and phrases indicating relationships or connections

No Comments

No Known Activations