INDEX

Explanations

concepts related to principles and doctrines

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_1/width_16k/average_l0_105

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 principles

-2.06

 principle

-2.02

 Principles

-1.88

principles

-1.84

principle

-1.79

Principles

-1.75

 Principle

-1.66

Principle

-1.55

 PRINCIPLES

-1.45

 principes

-1.30

POSITIVE LOGITS

abestanden

0.75

 للمعارف

0.71

setViewportView

0.70

!("{}",

0.69

consulté

0.69

 <>",

0.68

 ModelExpression

0.67

 насељу

0.67

testcase

0.66

titleMargin

0.65

Activations Density 0.164%

concepts related to principles and doctrines

No Comments

No Known Activations