INDEX

Explanations

references to self-related concepts and identities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

OGND

-0.74

 motivadoras

-0.62

endwhile

-0.61

 ofreci

-0.58

 Liturgy

-0.58

 borbo

-0.54

 élector

-0.54

новременно

-0.53

aspectj

-0.53

 nisso

-0.52

POSITIVE LOGITS

 AssemblyCulture

0.84

Vidite

0.66

AsUp

0.62

 localVar

0.61

 Kors

0.61

最快更新

0.60

 DBNull

0.59

RegressionTest

0.59

脚注の使い方

0.59

tanleria

0.58

Activations Density 0.086%

references to self-related concepts and identities

No Comments

No Known Activations