INDEX

Explanations

terms related to allegations of wrongdoing or misconduct

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

DECREF

-0.57

 Völker

-0.49

ñada

-0.46

inqu

-0.43

rah

-0.42

 disponib

-0.42

Fine

-0.42

 outputStream

-0.42

POSITIVE LOGITS

 alleges

1.19

 allege

1.14

 alleged

1.11

edly

1.08

 allegedly

1.08

ConstraintMaker

1.03

 Alleg

1.02

 alleging

0.99

Alleg

0.98

parsedMessage

0.92

Activations Density 0.259%

terms related to allegations of wrongdoing or misconduct

No Comments

No Known Activations