INDEX

Explanations

references to "dementia" and related 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-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

 Lingkungan

-0.71

rrggbb

-0.63

 ainfi

-0.62

 Utara

-0.60

 myſelf

-0.59

 preſent

-0.59

 Tinggi

-0.58

 régionale

-0.57

 Lainnya

-0.56

 régionales

-0.56

POSITIVE LOGITS

Dem

0.82

Deb

0.81

DEM

0.80

De

0.62

Dem

0.62

Deb

0.60

Dev

0.58

DEM

0.58

Del

0.56

 decom

0.56

Activations Density 0.012%

references to "dementia" and related terms

No Comments

No Known Activations