INDEX

Explanations

references to personal names or identifiers

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

 arşivlendi

-0.77

 disambiguazione

-0.71

endcsname

-0.63

seamnă

-0.56

enderror

-0.54

 endregion

-0.54

BufferException

-0.49

 verantwoorde

-0.48

__(/*!

-0.48



-0.47

POSITIVE LOGITS

 prénom

0.55

firstName

0.45

 firstname

0.42

rungsseite

0.42

 names

0.42

Nombres

0.41

names

0.40

 firstName

0.38

firstname

0.38

AsUp

0.38

Activations Density 0.003%

references to personal names or identifiers

No Comments

No Known Activations