INDEX

Explanations

phrases indicating relationships or connections between concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

enumi

-0.97

ViewFeatures

-0.78

に

-0.75

 Ballard

-0.75

Fakta

-0.73

 podjela

-0.72

""}

-0.71

()]

-0.71

";}

-0.70

 REED

-0.70

POSITIVE LOGITS

 vermelhas

0.74

 verksam

0.73

__*/

0.72

 fibres

0.72

 AssemblyTitle

0.70

 sjø

0.70

 tør

0.70

RectangleBorder

0.70

Of

0.70

 Nato

0.69

Activations Density 0.180%

phrases indicating relationships or connections between concepts

No Comments

No Known Activations