INDEX

Explanations

references to workouts and exercise sessions

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

 workout

-0.92

'])){

-0.71

scryfall

-0.68

cerely

-0.68

ently

-0.67

ificance

-0.66

rens

-0.66

YSTEM

-0.65

alised

-0.64

 workouts

-0.64

POSITIVE LOGITS

 vérit

0.65

shall

0.61

AppMethodBeat

0.58

Carrière

0.58

survi

0.56

Outputs

0.55

 eût

0.55

thető

0.55

 scolaires

0.54

 critères

0.54

Activations Density 0.008%

references to workouts and exercise sessions

No Comments

No Known Activations