INDEX

Explanations

lists and types of strategies or items in various contexts

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

terms related to languages and their classification or list

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

alphabetically sorted lists

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

 ostavi

-0.56

 pourtant

-0.55

expandindo

-0.54

ʺ

-0.51

HasForeignKey

-0.51

ESSIONAL

-0.50

httphttps

-0.48

CodeAttribute

-0.47

最快更新

-0.47

AndroidJUnit

-0.47

POSITIVE LOGITS

 beserta

1.07

及其

0.99

 alphabetically

0.95

 wraz

0.95

 their

0.95

their

0.89

Their

0.86

とその

0.82

 sorted

0.81

 along

0.80

Activations Density 0.267%

lists and types of strategies or items in various contexts

terms related to languages and their classification or list

alphabetically sorted lists

No Comments

No Known Activations