INDEX

Explanations

terms and phrases related to performance metrics and data organization in system components

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

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

 refroid

-0.57

Begriffsklä

-0.55

脚注の使い方

-0.54

Rüyada

-0.53

 obé

-0.51

 conséquence

-0.51

 يتيمه

-0.50

 Normdatei

-0.49

 căr

-0.49

onAttach

-0.48

POSITIVE LOGITS

 statistics

1.16

 stats

1.14

 statistic

1.04

 Stats

1.02

 Statistics

1.00

 metrics

1.00

stats

0.99

 total

0.96

统计

0.94

 averages

0.94

Activations Density 0.503%

terms and phrases related to performance metrics and data organization in system components

No Comments

No Known Activations