INDEX

Explanations

elements related to graph models and their properties

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 CanadaChoose

-0.68

ValueStyle

-0.66

 nakalista

-0.65

AddTagHelper

-0.63

Билгалдахарш

-0.63

ImageContext

-0.63

 &___

-0.59

+#+

-0.58

oa̍t

-0.58

 ProtoMessage

-0.58

POSITIVE LOGITS

 mixta

0.35

 classified

0.35

ระบบ

0.35

 system

0.34

 wakil

0.33

 rispondere

0.32

 Geheim

0.32

 institucional

0.32

 systems

0.31

 RoutedEventArgs

0.31

Activations Density 0.051%

elements related to graph models and their properties

No Comments

No Known Activations