INDEX

Explanations

instances of the word "regret" and references to gender demographics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/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.0.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

 ainfi

-0.77

 ilustracji

-0.67

colgante

-0.61

 satılık

-0.59

pecabe

-0.58

 valentín

-0.58

 venganza

-0.58

ambién

-0.57

 plufieurs

-0.56

 nadzie

-0.56

POSITIVE LOGITS

mod

0.58

bread

0.56

 prop

0.56

kna

0.56

 parap

0.55

 prob

0.54

 adven

0.53

 Samaria

0.53

ord

0.52

ORD

0.52

Activations Density 0.299%

instances of the word "regret" and references to gender demographics

No Comments

No Known Activations