INDEX

Explanations

elements and concepts related to revenge and the consequences of past actions

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

followed by prepositions

np_acts-logits-general · gemini-2.0-flash

qualities and concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

非常的

-0.67

 iaitu

-0.62

wußt

-0.62

/>";

-0.59

maktadır

-0.59

 todays

-0.58

凄く

-0.57

 didalam

-0.56

 entanto

-0.56

 disini

-0.56

POSITIVE LOGITS

 (“

1.21

("

0.97

 lest

0.89

—“

0.86

(!

0.81

 pronto

0.77

 courtesy

0.77

 chez

0.76

(“

0.76

(!)

0.75

Activations Density 1.098%

elements and concepts related to revenge and the consequences of past actions

followed by prepositions

qualities and concepts

No Comments

No Known Activations