INDEX

Explanations

the term "amen" and variations of it

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_0/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.hook_mlp_out

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

LookAnd

-0.85

 EnglishChoose

-0.82

 ModelExpression

-0.78

CodeAttribute

-0.71

arean

-0.69

IndentedString

-0.68

 "..\..\..\

-0.66

ThroughAttribute

-0.66

 calculées

-0.66

caloosa

-0.66

POSITIVE LOGITS

brainly

0.59

uestions

0.52

assertRaises

0.52

es

0.51

 sall

0.50

&=&

0.49

veless

0.49

 Unterschied

0.48

assertIs

0.48

Activations Density 0.062%

the term "amen" and variations of it

No Comments

No Known Activations