INDEX

Explanations

significant numerical data or parameters related to models and analysis

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

HTML tags and code constructs

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

ագրություններ

-0.82

 surla

-0.79

AndEndTag

-0.78

SequentialGroup

-0.71

usermodel

-0.71

niſſe

-0.70

ſicht

-0.69

 Geiſt

-0.68

majánló

-0.67

下载附件

-0.66

POSITIVE LOGITS

0.39

<u>

0.38

<h1>

0.36

↵↵↵

0.35

The

0.35

0.34

Pod

0.34

...

0.34

Da

0.34

<h4>

0.33

Activations Density 0.074%

significant numerical data or parameters related to models and analysis

HTML tags and code constructs

No Comments

No Known Activations