INDEX

Explanations

the letter "D" in various contexts

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

"D" followed by a lowercase letter or number

np_acts-logits-general · gemini-2.0-flash

D followed by certain 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

SOUNDBITE

-0.71

 kasarigan

-0.71

ata

-0.56

Referanser

-0.56

AddTagHelper

-0.48

oc

-0.48

ate

-0.48

ates

-0.46

Cyfeiriadau

-0.45

Referințe

-0.45

POSITIVE LOGITS

 Paglinawan

0.65

 propOrder

0.64

setVerticalGroup

0.56

 CreateTagHelper

0.55

 doInBackground

0.55

expandindo

0.54

 anledning

0.53

 vermelhas

0.51

 Gospels

0.51

 scattata

0.51

Activations Density 0.265%

the letter "D" in various contexts

"D" followed by a lowercase letter or number

D followed by certain characters

No Comments

No Known Activations