INDEX

Explanations

phrases indicating overall assessment or significant conclusions

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

пар

-0.53

hin

-0.51

 party

-0.51

pot

-0.51

geben

-0.50

 lenker

-0.50

pre

-0.50

ardy

-0.49

 stran

-0.48

POSITIVE LOGITS

 widespread

1.18

 pervasive

1.05

 prevalence

1.04

idespread

1.03

overall

1.03

 outright

0.96

 overall

0.95

 prevalent

0.94

 itſelf

0.90

 becauſe

0.83

Activations Density 0.118%

phrases indicating overall assessment or significant conclusions

No Comments

No Known Activations