INDEX

Explanations

words related to political and policy discussions

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

aky

-0.57

chuk

-0.56

 frontline

-0.54

 explan

-0.54

coni

-0.53

aga

-0.52

 Vaugh

-0.52

 proble

-0.51

iny

-0.50

[*

-0.48

POSITIVE LOGITS

huh

0.89

etc

0.88

 please

0.78

please

0.75

govtrack

0.72

channelAvailability

0.66

eh

0.64

 partName

0.62

flush

0.60

 plus

0.60

Activations Density 0.166%

words related to political and policy discussions

No Comments

No Known Activations

words related to political and policy discussions

No Comments

No Known Activations