INDEX

Explanations

proper nouns and names related to locations, people, and scientific terms

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

CppMethod

-0.61

bewerken

-0.61

 defaultstate

-0.59

">//

-0.58

دانشنامهٔ

-0.57

 <=",

-0.57

Personensuche

-0.57

 ſind

-0.56

Autoritní

-0.54

 Audiodateien

-0.54

POSITIVE LOGITS

ificantly

0.36

baarheid

0.33

AllowUser

0.32

Công

0.30

hal

0.30

хай

0.30

IInter

0.30

IDA

0.30

ida

0.30

</i>

0.29

Activations Density 0.349%

proper nouns and names related to locations, people, and scientific terms

No Comments

No Known Activations