INDEX

Explanations

words related to controversial or sensitive topics

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Ferdinand

-0.60

Downloadha

-0.57

 Spiel

-0.56

Eb

-0.55

 Teddy

-0.55

osaurs

-0.55

emonic

-0.54

Ou

-0.54

oulos

-0.54

zyk

-0.54

POSITIVE LOGITS

ellation

0.77

bara

0.71

Redd

0.69

ongyang

0.67

lege

0.66

 Thumbnails

0.66

Ĩ

0.63

shadow

0.62

eper

0.60

anka

0.59

Activations Density 0.100%

words related to controversial or sensitive topics

No Comments

No Known Activations