INDEX

Explanations

references to tables and sites

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

 Normdatei

-0.42

 '\\;'

-0.39

PDATE

-0.39

 kasarigan

-0.39

CppMethod

-0.37

SharedDtor

-0.36

OGND

-0.36

 appointment

-0.36

 delicately

-0.36

atschappij

-0.35

POSITIVE LOGITS

 tables

0.71

 site

0.70

site

0.70

 Tables

0.60

 comfy

0.60

 дописавши

0.60

 oral

0.59

CPtr

0.58

 sites

0.56

Tables

0.56

Activations Density 0.238%

references to tables and sites

No Comments

No Known Activations