INDEX

Explanations

keywords related to written documents and research studies

oai_token-act-pair · gpt-3.5-turbo

words related to scholarly articles and documentation

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

 forbid

-0.65

inyl

-0.65

emia

-0.64

erald

-0.60

icans

-0.60

 brakes

-0.60

rogens

-0.58

 crucifix

-0.57

ographers

-0.57

 bastard

-0.57

POSITIVE LOGITS

æĪ¦

0.72

emp

0.68

ç¥ŀ

0.66

 guise

0.66

ĨĴ

0.66

ahime

0.64

 shootout

0.64

quad

0.63

session

0.62

stakes

0.62

Activations Density 0.526%

keywords related to written documents and research studies

words related to scholarly articles and documentation

No Comments

No Known Activations