INDEX

Explanations

references to television shows and characters from the CW network

oai_token-act-pair · gpt-3.5-turbo

references to the CW network and its shows

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

urers

-0.85

zsche

-0.79

ãĥĥãĥĪ

-0.76

quez

-0.71

¬¼

-0.70

interstitial

-0.68

Ń·

-0.68

insula

-0.67

ogun

-0.66

eger

-0.66

POSITIVE LOGITS

lev

0.72

Rating

0.70

 Zone

0.68

Zone

0.68

INGTON

0.67

lett

0.67

Stand

0.67

abit

0.65

Recomm

0.65

trop

0.63

Activations Density 0.054%

references to television shows and characters from the CW network

references to the CW network and its shows

No Comments

No Known Activations