INDEX

Explanations

references to variables and their corresponding values in programming or scientific contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

clero

-0.52

er

-0.46

 Longo

-0.44

localctx

-0.44

茜

-0.44

princip

-0.43

 Infórmanos

-0.43

Đi

-0.42

 martyrs

-0.42

POSITIVE LOGITS

vv

0.84

vav

0.79

0.78

VIAF

0.76

vula

0.72

 AssemblyTitle

0.69

iv

0.69

vă

0.69

vy

0.68

MAV

0.67

Activations Density 0.743%

references to variables and their corresponding values in programming or scientific contexts

No Comments

No Known Activations