INDEX

Explanations

specific product or brand names in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

LEncoder

-0.63

 Shakspeare

-0.59

HasForeignKey

-0.57

findpost

-0.56

 photolibrary

-0.54

 oakley

-0.53

 SNCF

-0.51

 geodes

-0.50

Dn

-0.50

 Thra

-0.50

POSITIVE LOGITS

 autorytatywna

0.73

 nahilalakip

0.73

ьаж

0.66

+#+#

0.66

Παραπομπές

0.63

 >=",

0.62

 ExecuteAsync

0.62

 ujednoznacz

0.60

featureID

0.60

Activations Density 0.115%

specific product or brand names in various contexts

No Comments

No Known Activations