INDEX

Explanations

terms related to mathematical or computational problems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.hook_resid_post

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

 betrekking

-0.60

 idéia

-0.59

 gavetas

-0.57

 Forderung

-0.53

 Erstellung

-0.53

 Aufwand

-0.53

 sobą

-0.53

 Begründung

-0.52

 wspar

-0.52

 ciepła

-0.50

POSITIVE LOGITS

 Proble

0.95

 Probl

0.90

 proble

0.86

 Problem

0.84

 problem

0.83

problem

0.81

proble

0.80

Problem

0.79

Proble

0.79

 buzz

0.79

Activations Density 0.217%

terms related to mathematical or computational problems

No Comments

No Known Activations