INDEX

Explanations

structured data and experimental results represented by figures and tables

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

CSD

-0.57

DST

-0.57

 Galla

-0.54

STC

-0.53

 Plin

-0.52

PCC

-0.52

BOC

-0.52

MMS

-0.51

KBS

-0.50

PSM

-0.50

POSITIVE LOGITS

 Wheatley

0.77

 McGowan

0.68

 Huerta

0.68

 Schäfer

0.64

 Henley

0.64

 Buckley

0.63

 McGill

0.63

 Hamza

0.61

 Simone

0.59

 Leverkusen

0.58

Activations Density 0.934%

structured data and experimental results represented by figures and tables

No Comments

No Known Activations