INDEX

Explanations

namespaces and import statements commonly found in programming and markup languages

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

Code snippets or XML tags, within text

np_acts-logits-general · gemini-2.0-flash

XML/Java/biological terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

"""

-0.34

<blockquote>

-0.29

 Boletín

-0.29

 resumen

-0.28

 Мексичка

-0.27

提

-0.27

/*++

-0.26

 promedio

-0.26

 tremp

-0.25

Démographie

-0.25

POSITIVE LOGITS

 colspan

2.67

 rowspan

1.42

 nahilalakip

1.00

 FontWeight

0.94

 GenerationType

0.83

multicolumn

0.81

########.

0.80

<unused41>

0.79

<unused28>

0.79

[@BOS@]

0.79

Activations Density 0.045%

namespaces and import statements commonly found in programming and markup languages

Code snippets or XML tags, within text

XML/Java/biological terms

No Comments

No Known Activations