INDEX

Explanations

assertive phrases that indicate certainty or confirmation

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

shown to be

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

PhysRevD

-0.52

writeFieldEnd

-0.49

 avoient

-0.49

 worse

-0.47

ViewImports

-0.47

StoryboardSegue

-0.46

 peores

-0.46

 PeEnEo

-0.45

 confusión

-0.45

 fisuras

-0.45

POSITIVE LOGITS

TestingModule

0.48

 mergeFrom

0.42

ябре

0.40

を務

0.40

 kasarigan

0.39

Rujuakan

0.38

IndentedString

0.38

ptos

0.37

Бележки

0.37

पया

0.37

Activations Density 0.083%

assertive phrases that indicate certainty or confirmation

shown to be

No Comments

No Known Activations