INDEX

Explanations

phrases indicating the best practices or optimal strategies in various contexts

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

optimal ways

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 unless

-0.33

ExecuteReader

-0.30

 also

-0.28

 ingår

-0.28

 malade

-0.28

 without

-0.28

 because

-0.27

unless

-0.27

TEntity

-0.27

 også

-0.27

POSITIVE LOGITS

 optimum

0.70

ंदीखरीदारी

0.67

 optimal

0.66

PREFERRED

0.64

ftagPool

0.63

хьтан

0.63

 Optimum

0.62

optimal

0.62

دانشنامهٔ

0.62

IndentedString

0.60

Activations Density 0.047%

phrases indicating the best practices or optimal strategies in various contexts

optimal ways

No Comments

No Known Activations