INDEX

Explanations

references to previously mentioned information or sections in the text

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

references to prior evidence or information in legal contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_0/width_16k/average_l0_119

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.hook_mlp_out

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

 AssemblyCompany

-0.69

Tikang

-0.56

 rispar

-0.56

 ImportError

-0.54

Personensuche

-0.54

 lowest

-0.54

 satunya

-0.54

czuk

-0.54

 Савезне

-0.54

 scales

-0.53

POSITIVE LOGITS

setupUi

0.63

 following

0.63

following

0.60

Diwedd

0.58

bufio

0.56

Whilst

0.56

WebVitals

0.55

urlpatterns

0.55

__':

0.55

andExpect

0.55

Activations Density 0.326%

references to previously mentioned information or sections in the text

references to prior evidence or information in legal contexts

No Comments

No Known Activations