INDEX

Explanations

phrases related to expressing doubt or uncertainty

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_fs768-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

768

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

PU

-0.66

 guided

-0.65

 protected

-0.64

 Mechdragon

-0.64

 nearest

-0.62

 dust

-0.60

 Anarchy

-0.59

 couch

-0.59

pus

-0.59

Adv

-0.58

POSITIVE LOGITS

't

1.76

ÃŃ

1.14

ned

1.13

uts

1.03

eness

1.02

itely

0.99

ates

0.98

Â´

0.98

iting

0.97

ited

0.94

Activations Density 2.761%

phrases related to expressing doubt or uncertainty

No Comments

No Known Activations