INDEX

Explanations

phrases indicating sources of information and the attribution of statements

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

cloudf

-0.55

rück

-0.54

AnchorTagHelper

-0.53

Datuak

-0.51

 Baldwin

-0.50

zzar

-0.49

tym

-0.49

 pageNo

-0.49

hofer

-0.49

romi

-0.48

POSITIVE LOGITS

NewUrlParser

0.67

日閲覧

0.63

ंदीखरीदारी

0.57

ArrowToggle

0.57

($__

0.54

----</

0.53

 sources

0.52

 interviewed

0.52

itinéraire

0.52

saraba

0.51

Activations Density 0.212%

phrases indicating sources of information and the attribution of statements

No Comments

No Known Activations