INDEX

Explanations

programming-related terms or code snippets

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

 Infórmanos

-0.69

abestanden

-0.66

MLLoader

-0.62

 تضيفلها

-0.62

 يتيمه

-0.58

 heav

-0.55

 Independent

-0.54

liflower

-0.54

 Tapatalk

-0.52

 '\\;'

-0.52

POSITIVE LOGITS

richTextPanel

0.57

usermodel

0.56

setDo

0.53

 Spisak

0.50

ništvo

0.50

 pés

0.49

 Савез

0.49

bè

0.49

IContainer

0.49

"{\

0.49

Activations Density 0.294%

programming-related terms or code snippets

No Comments

No Known Activations