INDEX

Explanations

references or mentions of "paper" and related terminology

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 paper

-3.06

paper

-2.89

 Paper

-2.77

Paper

-2.72

 PAPER

-2.45

PAPER

-2.28

 papers

-2.08

papers

-1.95

 Papers

-1.91

Papers

-1.75

POSITIVE LOGITS

fert

0.46

 лице

0.45

agaan

0.43

ostock

0.43

createObject

0.43

BagLayout

0.42

 candidats

0.40

 colectiva

0.40

ureusement

0.40

manteau

0.40

Activations Density 0.188%

references or mentions of "paper" and related terminology

No Comments

No Known Activations