INDEX

Explanations

names of individuals and elements related to different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

parsedMessage

-0.85

OGND

-0.82

expandindo

-0.75

RenderAtEndOf

-0.70

 BoxDecoration

-0.69

TestingModule

-0.69

 صوتيه

-0.67

IUrlHelper

-0.66

GOTREF

-0.65

CppMethod

-0.65

POSITIVE LOGITS

and

0.33

0.30

 dispon

0.29

 идет

0.28

or

0.28

 import

0.27

 же

0.27

and

0.27

kolle

0.27

oo

0.27

Activations Density 1.895%

names of individuals and elements related to different contexts

No Comments

No Known Activations