INDEX

Explanations

mathematical expressions and syntax

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

 kasarigan

-0.63

awtextra

-0.50

Autoritní

-0.46

ViewFeatures

-0.45

mstyle

-0.42

SOUNDBITE

-0.41

ThroughAttribute

-0.40

RTSN

-0.39

teardown

-0.38

 sobě

-0.38

POSITIVE LOGITS

 ruinas

0.38

his

0.37

iettivo

0.36

 Olímpicos

0.36

EndContext

0.36

 weak

0.36

rboles

0.36

 neither

0.35

паси

0.35

bitField

0.35

Activations Density 0.278%

mathematical expressions and syntax

No Comments

No Known Activations