INDEX

Explanations

instances of legal modifications or corrections

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

modified

-0.76

 modified

-0.60

addPreferredGap

-0.56

dasarkan

-0.54

hicle

-0.54

edah

-0.54

äksi

-0.53

 resultar

-0.53

Sa

-0.52

POSITIVE LOGITS

:✨

0.81

 CanadaChoose

0.80

NameInMap

0.75

 تكبرها

0.74

fml

0.73

 createSlice

0.73

PerformLayout

0.71

parsedMessage

0.71

 TestBed

0.71

 Familienname

0.70

Activations Density 0.019%

instances of legal modifications or corrections

No Comments

No Known Activations