INDEX

Explanations

sentence structures and punctuation that suggest clarity and conclusion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

المناصب

-0.56

 autorytatywna

-0.53

Dic

-0.52

lectic

-0.49

évaluateur

-0.48

 humming

-0.48

 Passes

-0.48

 Else

-0.47

aluc

-0.47

ViewSet

-0.47

POSITIVE LOGITS

 shocking

0.47

 shocked

0.44

 coupable

0.42

 terke

0.41

 sorpresa

0.41

 énergétique

0.40

 surpresa

0.40

 descubrir

0.39

 StatelessWidget

0.39

BeginInit

0.39

Activations Density 0.058%

sentence structures and punctuation that suggest clarity and conclusion

No Comments

No Known Activations