INDEX

Explanations

terms related to competition and rivalries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

Lec

-0.66

 condition

-0.60

Personendaten

-0.58

zab

-0.55

providedIn

-0.53

 conditions

-0.53

ras

-0.53

conditions

-0.53

出版年

-0.52

dö

-0.52

POSITIVE LOGITS

 competing

1.93

 rival

1.79

 competes

1.70

 compete

1.59

 rivals

1.57

 competed

1.54

Rival

1.51

Competing

1.51

 Rival

1.40

rival

1.38

Activations Density 0.113%

terms related to competition and rivalries

No Comments

No Known Activations