INDEX

Explanations

references to authority and control in the context of societal issues

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

 acquaintances

-0.66

 overflow

-0.62

 beautifully

-0.61

 briefly

-0.60

 aliases

-0.59

 unsuccessfully

-0.58

 linem

-0.58

 contemporaries

-0.58

 renown

-0.58

asionally

-0.58

POSITIVE LOGITS

 Same

0.69

 footing

0.69

nuclear

0.68

osate

0.67

same

0.65

 brunt

0.64

farious

0.63

ĪĴ

0.63

 blame

0.62

ocratic

0.62

Activations Density 0.176%

references to authority and control in the context of societal issues

No Comments

No Known Activations