INDEX

Explanations

terms and phrases related to medical advisories and company leadership

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

steps in proofs and procedures

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.24.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

TypedDataSet

-0.70

 estekak

-0.65

MigrationBuilder

-0.64

">//

-0.60

 &___

-0.59

 betweenstory

-0.56

 виправивши

-0.55

oredCriteria

-0.54

 برانيه

-0.52

 ProtoMessage

-0.51

POSITIVE LOGITS

UrlResolution

0.48

Datuak

0.44

ogno

0.41

 EClass

0.40

VIAF

0.37

Filmografie

0.36

XmlAccessorType

0.36

WriteTagHelper

0.36

 Joseph

0.35

WebElementEntity

0.35

Activations Density 0.276%

terms and phrases related to medical advisories and company leadership

steps in proofs and procedures

No Comments

No Known Activations