INDEX

Explanations

the word "Se" and its variants in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 nahilalakip

-0.75

Datuak

-0.71

 Houſe

-0.70

")==

-0.70

FTFY

-0.68

hastly

-0.67

awtextra

-0.66

convenience

-0.66

 imper

-0.65

\{\\

-0.65

POSITIVE LOGITS

Se

1.09

Se

0.87

 aggressively

0.58

 aggressive

0.51

Portale

0.51

 segregation

0.51

 aggressiveness

0.50

nm

0.50

 unemployment

0.50

ranath

0.50

Activations Density 0.082%

the word "Se" and its variants in different contexts

No Comments

No Known Activations