INDEX

Explanations

prepositions, particularly focusing on "for."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

)";

-0.75

fromnode

-0.72

>';

-0.67

ukone

-0.65

makeText

-0.64

>";

-0.62

MemoryWarning

-0.60

Ārējās

-0.59

 TestBed

-0.58

)";

-0.58

POSITIVE LOGITS

Για

0.74

for

0.74

for

0.73

 purposes

0.72

 für

0.70

für

0.68

 sake

0.66

FOR

0.66

FOR

0.66

addContainerGap

0.66

Activations Density 0.281%

prepositions, particularly focusing on "for."

No Comments

No Known Activations