INDEX

Explanations

references to datasets

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

 propOrder

-0.72

 miei

-0.65

ьаж

-0.65

Enumerated

-0.63

InSection

-0.59

 disambiguazione

-0.58

 surla

-0.58

рованы

-0.58

Morfologia

-0.57

Ầ

-0.57

POSITIVE LOGITS

enumi

0.73

enumii

0.67

 purpoſe

0.66

 contextLoads

0.65

 الحره

0.65

hairs

0.63

{}".

0.61

GraphicsUnit

0.61

ſelf

0.61

yntaxException

0.60

Activations Density 0.083%

references to datasets

No Comments

No Known Activations