INDEX

Explanations

online profiles and blog entries

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

 appeal

-0.68

 Lung

-0.62

 memor

-0.60

 representation

-0.59

 foundation

-0.58

 portray

-0.58

 pursu

-0.58

 rejection

-0.57

 representations

-0.56

 elements

-0.55

POSITIVE LOGITS

 Joined

1.21

Joined

0.91

Offline

0.91

 Profile

0.88

 Member

0.87

 Blog

0.83

à¼

0.80

ð

0.79

join

0.79

Activations Density 0.147%

online profiles and blog entries

No Comments

No Known Activations