INDEX

Explanations

names and references to political figures and parties

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

 كومونز

-0.63

 الرياضيه

-0.60

StoreMessageInfo

-0.57

bootstrapcdn

-0.54

 beginnetje

-0.53

 Maharaja

-0.52

 Texans

-0.52

parsedMessage

-0.52

messageInfo

-0.52

MigrationBuilder

-0.51

POSITIVE LOGITS

 reportedly

0.78

lieri

0.57

 allegedly

0.56

 ostensibly

0.56

 purported

0.56

 supuestamente

0.55

ourite

0.54

 Donahue

0.54

 supposedly

0.54

 unidentified

0.54

Activations Density 0.127%

names and references to political figures and parties

No Comments

No Known Activations