INDEX

Explanations

statements and claims made by individuals regarding their beliefs or actions

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

verbs of saying and reporting

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

oredCriteria

-0.54

原始内容存档于

-0.53

 arşivlendi

-0.49

GraphicsUnit

-0.45

glLoadIdentity

-0.44

 AssemblyCompany

-0.44

 AssemblyCulture

-0.43

 Administrativna

-0.42

wahati

-0.42

 snippetHide

-0.42

POSITIVE LOGITS

 gracia

0.49

 claims

0.46

CLAIM

0.42

 bragging

0.41

的说

0.41

 refers

0.40

Vezi

0.40

lists

0.40

 errado

0.40

lagem

0.40

Activations Density 0.132%

statements and claims made by individuals regarding their beliefs or actions

verbs of saying and reporting

No Comments

No Known Activations