INDEX

Explanations

references to famous locations or landmarks

oai_token-act-pair · gpt-3.5-turbo

phrases related to popular culture and significant events

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

 (âĪĴ

-0.70

 Intermediate

-0.59

Newsletter

-0.59

 complying

-0.59

 administr

-0.56

exit

-0.56

 Ibid

-0.55

keeping

-0.55

.–

-0.55

 calculation

-0.54

POSITIVE LOGITS

 hilar

0.71

 comedic

0.68

 thrill

0.68

 classy

0.68

 genre

0.67

 Comedy

0.67

sci

0.67

 hilarious

0.66

 comedy

0.65

 nerd

0.63

Activations Density 2.015%

references to famous locations or landmarks

phrases related to popular culture and significant events

No Comments

No Known Activations