INDEX

Explanations

phrases related to political discussion and controversial 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

erves

-0.74

=-=-=-=-

-0.68

rehens

-0.68

 nevertheless

-0.68

_-

-0.65

 furthermore

-0.64

ÃŃs

-0.64

ATCH

-0.63

 nonetheless

-0.62

Ö¼

-0.61

POSITIVE LOGITS

judicial

0.89

"#

0.88

ordinary

0.85

 caliphate

0.82

0.80

social

0.80

metadata

0.79

 Islamic

0.79

advant

0.73

prime

0.69

Activations Density 0.023%

phrases related to political discussion and controversial topics

No Comments

No Known Activations