INDEX

Explanations

phrases starting with "none of" or "none," indicating negation or the absence of something

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

SequentialGroup

-0.62

NameInMap

-0.58

MemoryWarning

-0.56

AISSEE

-0.54

tanleria

-0.54

 sogen

-0.52

AppRoutingModule

-0.52

CPC

-0.52

InstrumentedTest

-0.52

лтамалар

-0.50

POSITIVE LOGITS

any

0.63

 ninguno

0.58

 none

0.57

 neither

0.56

 ninguna

0.55

None

0.53

none

0.53

 nessuna

0.53

 whatsoever

0.52

 NONE

0.52

Activations Density 0.011%

phrases starting with "none of" or "none," indicating negation or the absence of something

No Comments

No Known Activations