INDEX

Explanations

instances of detailed narration or descriptions in texts

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

featureID

-0.80

ſelf

-0.63

ſtanding

-0.61

ftagPool

-0.61

 queſta

-0.60

 NSCoder

-0.59

 Jefus

-0.59

GEBURTSDATUM

-0.59

astéroïdes

-0.59

-0.58

POSITIVE LOGITS

 Gesch

0.35

nicy

0.34

 courtesy

0.32

 लग

0.31

้า

0.31

 Weit

0.31

AutoScaleMode

0.30

 parte

0.30

 done

0.29

Sch

0.29

Activations Density 0.998%

instances of detailed narration or descriptions in texts

No Comments

No Known Activations