INDEX

Explanations

elements related to function definitions and code structure

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

 Engl

-0.46

niku

-0.43

hep

-0.40

يل

-0.40

boy

-0.40

 tazas

-0.39

vil

-0.39

िला

-0.39

apan

-0.39

Raise

-0.39

POSITIVE LOGITS

Tikang

0.97

 disambiguazione

0.79

 विश्वसनीयता

0.78

expandindo

0.70

kegaard

0.68

 initComponents

0.68

rungsseite

0.68

BeginContext

0.67

 himo

0.66

ंदीखरीदारी

0.66

Activations Density 0.060%

elements related to function definitions and code structure

No Comments

No Known Activations