INDEX

Explanations

frequent terms related to common practices or standards in various contexts

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

HomeAsUpEnabled

-0.66

parsedMessage

-0.65

 Superhost

-0.60

Glej

-0.60

 researched

-0.58

räck

-0.58

Research

-0.57

)))

-0.57

 igenom

-0.56

 researches

-0.55

POSITIVE LOGITS

HasAnnotation

0.70

titleMargin

0.65

 Roskov

0.60

กัน

0.59

AccessFile

0.57

UIApplication

0.57

 fontaine

0.53

maxLength

0.51

 until

0.51

GeneratedMessage

0.51

Activations Density 0.518%

frequent terms related to common practices or standards in various contexts

No Comments

No Known Activations