INDEX

Explanations

phrases indicating identification or classification of subjects

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

identified as

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

évaluateur

-0.49

Containing

-0.43

 tucked

-0.43

perg

-0.42

Tikang

-0.39

InSection

-0.38

 Antal

-0.37

tanleria

-0.37

stars

-0.37

MathML

-0.37

POSITIVE LOGITS

 <=",

0.63

msgTypes

0.48

 identified

0.46

 XCTest

0.46

 GetEnumerator

0.45

 ویکی‌پدی

0.45

 identifiés

0.45

AccessorTable

0.44

يكب

0.43

UpInside

0.42

Activations Density 0.000%

phrases indicating identification or classification of subjects

identified as

No Comments

No Known Activations

phrases indicating identification or classification of subjects

identified as

No Comments

No Known Activations