INDEX

Explanations

references to penalties, particularly in legal contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

cplusplus

-0.44

posições

-0.42

勤

-0.39

 Employment

-0.38

 exer

-0.38

めで

-0.37

 АК

-0.37

tisgarh

-0.37

 warung

-0.37

requent

-0.37

POSITIVE LOGITS

 AssemblyTitle

0.69

AccessorTable

0.64

 AssemblyCompany

0.60

ThroughAttribute

0.60

ThemeOverlay

0.59

 ObjectParameter

0.57

providedIn

0.56

StartTag

0.55

autorest

0.55

DH

0.54

Activations Density 0.359%

references to penalties, particularly in legal contexts

No Comments

No Known Activations