INDEX

Explanations

terms related to multifunctionality and integration in various products or solutions

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

CopyWith

-0.46

netinet

-0.41

\{\\

-0.40

 homophobic

-0.36

antía

-0.36

akana

-0.35

ienna

-0.35

setPower

-0.35

CheckBreak

-0.34

 }^{[

-0.34

POSITIVE LOGITS

 pleaſure

0.61

 multifunctional

0.61

 onOptions

0.60

 multipurpose

0.58

 houſe

0.54

 myſelf

0.54

 tartalomajánló

0.51

 themſelves

0.48

combo

0.48

 uſe

0.48

Activations Density 0.014%

terms related to multifunctionality and integration in various products or solutions

No Comments

No Known Activations