INDEX

Explanations

references to categorizations or rankings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 both

-0.52

Both

-0.50

 Both

-0.48

Two

-0.44

om

-0.44

two

-0.44

BOTH

-0.43

mer

-0.42

あた

-0.42

Ambos

-0.42

POSITIVE LOGITS

ValueStyle

1.04

 Савезне

1.02

ArrowToggle

0.98

)");

0.96

 Roskov

0.95

 للاسماء

0.93

RegressionTest

0.91

IntoConstraints

0.90

 autorytatywna

0.89



0.87

Activations Density 0.939%

references to categorizations or rankings

No Comments

No Known Activations