INDEX

Explanations

numerical values and measurements that imply size and capacity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

+#+#

-0.64

enschappelijke

-0.61

ArgsConstructor

-0.61

 مرئيه

-0.60

 spokes

-0.59

unmodifiable

-0.58

 kasarigan

-0.57

Suara

-0.55

ruppen

-0.55

 ralla

-0.54

POSITIVE LOGITS

DeleteBehavior

0.53

 average

0.50

 dalamnya

0.49

 completos

0.49

hilangan

0.47

Gemeinden

0.47

 disparu

0.47

 équi

0.47

 normalt

0.47

单个

0.46

Activations Density 0.301%

numerical values and measurements that imply size and capacity

No Comments

No Known Activations