INDEX

Explanations

terms related to dominance and power dynamics within various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 escar

-0.48

 Scienti

-0.46

 Freiw

-0.43

testdata

-0.41

songwriter

-0.41

 Alcott

-0.41

 FBref

-0.41

rdata

-0.41

Sparks

-0.41

Funds

-0.40

POSITIVE LOGITS

 dominance

0.96

 domination

0.86

 Domin

0.84

Domin

0.82

 domínio

0.82

domin

0.82

 dominated

0.81

 dominant

0.81

 dominate

0.80

 dominating

0.80

Activations Density 0.472%

terms related to dominance and power dynamics within various contexts

No Comments

No Known Activations