INDEX

Explanations

terms related to negative sentiments or non-ideal situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.36.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

webElementXpaths

-0.52

 ChromeDriver

-0.51

 Photocase

-0.49

 ContentValues

-0.49

featureID

-0.48

 Normdatei

-0.46

PRNewswire

-0.46

MENAFN

-0.44

chieht

-0.43

 naselje

-0.43

POSITIVE LOGITS

usammen

0.48

ubs

0.43

ub

0.41

 pace

0.40

 Bekasi

0.40

 McColl

0.38

cin

0.38

بخ

0.38

atur

0.37

Activations Density 0.767%

terms related to negative sentiments or non-ideal situations

No Comments

No Known Activations