INDEX

Explanations

quantitative references or estimates in a context related to memories and historical reflection

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

many and most

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

and

-0.43

-0.38

 επί

-0.37

 собой

-0.37

staan

-0.36

بخش

-0.36

ManagedBean

-0.35

DllImport

-0.35

ilä

-0.35

اذ

-0.35

POSITIVE LOGITS

 defaultstate

0.91

aarrggbb

0.88

 تضيفلها

0.85

GraphicsUnit

0.84

 others

0.83

principalTable

0.82

 nahilalakip

0.81

Others

0.78

 חיצוניים

0.76

 ervan

0.75

Activations Density 0.093%

quantitative references or estimates in a context related to memories and historical reflection

many and most

No Comments

No Known Activations