INDEX

Explanations

information related to media and entertainment, specifically focusing on actor roles and industry news

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

chio

-0.75

ãħĭ

-0.72

alogue

-0.69

wcs

-0.67

thur

-0.65

mber

-0.64

vic

-0.62

acion

-0.62

 guiActiveUn

-0.60

 Guilty

-0.60

POSITIVE LOGITS

assetsadobe

1.11

0.76

 Macy

0.68

scenes

0.64

iday

0.62

inton

0.62

inki

0.61

igans

0.61

igan

0.61

Activations Density 0.034%

information related to media and entertainment, specifically focusing on actor roles and industry news

No Comments

No Known Activations