INDEX

Explanations

references to individuals or groups of people

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

"who" followed by a verb

np_acts-logits-general · gemini-2.0-flash

who followed by action

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

///</

-0.81

 snowing

-0.69

 رہ

-0.69

 internetowa

-0.69

 Walkover

-0.63

)]);

-0.60

ArrowToggle

-0.60

//}

-0.60

expandindo

-0.59

 circulation

-0.59

POSITIVE LOGITS

who

1.16

Who

0.92

who

0.86

Who

0.85

WHO

0.84

oping

0.77

 whom

0.75

Whoosh

0.71

 else

0.70

Whoo

0.68

Activations Density 0.083%

references to individuals or groups of people

"who" followed by a verb

who followed by action

No Comments

No Known Activations