INDEX

Explanations

words related to computer programming, technical optimization, and software demos

oai_token-act-pair · gpt-3.5-turbo

elements related to notation and formatting in text

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

EStream

-0.85

ortium

-0.80

wagen

-0.78

 conduc

-0.74

usalem

-0.72

wright

-0.71

aea

-0.70

 warmed

-0.67

earchers

-0.66

 sadly

-0.65

POSITIVE LOGITS

sounding

0.96

catentry

0.83

enough

0.74

aggressive

0.72

ities

0.69

DATA

0.68

SOURCE

0.67

VALUE

0.67

adj

0.67

yet

0.67

Activations Density 0.474%

words related to computer programming, technical optimization, and software demos

elements related to notation and formatting in text

No Comments

No Known Activations