INDEX

Explanations

text related to JavaScript programming

oai_token-act-pair · gpt-3.5-turbo

data structure indicators, specifically related to JSON or similar formats

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

yip

-0.74

istance

-0.68

icut

-0.68

regulated

-0.66

abb

-0.66

apo

-0.65

ifiable

-0.65

icate

-0.64

owl

-0.63

worst

-0.62

POSITIVE LOGITS

ously

0.95

ÏĦ

0.81

andum

0.81

AGE

0.79

vation

0.78

Ïĥ

0.75

opsis

0.73

OUS

0.73

âĸ¬

0.71

opol

0.70

Activations Density 0.036%

text related to JavaScript programming

data structure indicators, specifically related to JSON or similar formats

No Comments

No Known Activations