INDEX

Explanations

proper nouns, particularly names of people, organizations, and places

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

company names and abbreviations

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

Ծանոթ

-0.56

 typelib

-0.49

tvguidetime

-0.47

endphp

-0.45

☚

-0.44

localctx

-0.44

Референце

-0.43

----</

-0.42

ACTERS

-0.42

 Exacts

-0.41

POSITIVE LOGITS

xsi

0.40

 directorio

0.40

PYX

0.39

 banques

0.38

 getEmail

0.38

Tikang

0.38

getKeyCode

0.36

 acús

0.35

 roleId

0.35

جمعیت

0.35

Activations Density 0.031%

proper nouns, particularly names of people, organizations, and places

company names and abbreviations

No Comments

No Known Activations