INDEX

Explanations

quantitative data or numerical values

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

Numbers

np_acts-logits-general · gemini-2.0-flash

long s characters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

-1.13

-1.08

-1.03

to

-0.86

in

-0.83

or

-0.82

for

-0.81

-0.80

POSITIVE LOGITS

 itſelf

1.64

 myſelf

1.61

 doubtnut

1.54

Efq

1.53

 Majefty

1.53

 Monfieur

1.51

 Jefus

1.50

 pleaſure

1.49

 Reſ

1.48

 purpoſe

1.46

Activations Density 0.656%

quantitative data or numerical values

Numbers

long s characters

No Comments

No Known Activations