INDEX

Explanations

expressions of uncertainty or memory recall

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

"I" followed by memory/recall words

np_acts-logits-general · gemini-2.0-flash

I remember, think, or don't know

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

 muualla

-0.49

ertale

-0.48

makeText

-0.46

 Wallpaper

-0.45

glEnd

-0.43

modb

-0.40

odu

-0.39

ods

-0.39

чан

-0.38

adura

-0.38

POSITIVE LOGITS

 memory

1.54

 recollection

1.46

 remember

1.39

 recall

1.34

memory

1.32

 remembers

1.31

 mémoire

1.27

 remembering

1.24

 recuerdo

1.23

remember

1.21

Activations Density 0.205%

expressions of uncertainty or memory recall

"I" followed by memory/recall words

I remember, think, or don't know

No Comments

No Known Activations