INDEX

Explanations

high values indicating quantities or scores in data outputs

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

repeated mentions or references to the API and its functionality in a programming context

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

archaic pronouns and titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.20.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

RenderAtEndOf

-0.57

">//

-0.54

три

-0.49

pola

-0.48

 noDo

-0.47

三重

-0.47

tr

-0.46

-0.45

تقاوى

-0.45

jak

-0.43

POSITIVE LOGITS

 myſelf

1.06

 itſelf

0.99

 Monfieur

0.94

 himſelf

0.89

 auffi

0.85

 ſche

0.85

 ſtate

0.84

 fevere

0.84

 fubject

0.83

 Majefty

0.83

Activations Density 0.007%

high values indicating quantities or scores in data outputs

repeated mentions or references to the API and its functionality in a programming context

archaic pronouns and titles

No Comments

No Known Activations