INDEX

Explanations

references to toolbars or toolbar-related elements in applications

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<em>

-0.50

-0.45

-0.43

-0.42

<strong>

-0.42

%")

-0.42

<h2>

-0.41

Check

-0.40

is

-0.40

<i>

-0.40

POSITIVE LOGITS

 Toolbar

1.47

toolbar

1.44

 toolbar

1.41

Toolbar

1.34

ToolBar

1.02

TOOLBAR

0.89

toolBar

0.77

 StringTokenizer

0.75

orologio

0.74

 TOOL

0.71

Activations Density 0.002%

references to toolbars or toolbar-related elements in applications

No Comments

No Known Activations