INDEX

Explanations

references to months in 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 @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

ListItemIcon

-0.44

hadapi

-0.44

aic

-0.43

printStackTrace

-0.43

 białe

-0.42

aeros

-0.42

laştır

-0.41

 isolado

-0.41

 Bantu

-0.41

 Piro

-0.41

POSITIVE LOGITS

 Month

1.09

Month

1.05

 month

1.02

 MONTH

0.96

 months

0.94

 Months

0.91

month

0.91

Months

0.86

MONTH

0.85

months

0.81

Activations Density 0.091%

references to months in various contexts

No Known Activations