INDEX

Explanations

references to numerical data and counts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.37.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

 habet

-0.43

 niyang

-0.41

ujud

-0.41

hæng

-0.41

mmä

-0.40

 võimal

-0.40

værende

-0.40

 moviliz

-0.40

 zufolge

-0.40

zbęd

-0.40

POSITIVE LOGITS

’

0.83

 <<<<<<<<<<<<<<

0.77

0.73

ViewImports

0.71

\'

0.61

em

0.60



0.60

 AssemblyTitle

0.59

0.56

 &___

0.55

Activations Density 0.324%

references to numerical data and counts

No Comments

No Known Activations

references to numerical data and counts

No Comments

No Known Activations