INDEX

Explanations

commands or suggestions related to user actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 TextInputType

-0.58

 testSet

-0.47

tably

-0.46

 Tapatalk

-0.45

прочем

-0.45

();)

-0.45

 loob

-0.44

 testGet

-0.43

atd

-0.43

 derselbe

-0.43

POSITIVE LOGITS

 THEN

0.68

contentLoaded

0.67

×

0.64

 dann

0.63

 then

0.60

oredCriteria

0.60

arrant

0.60

umenter

0.57

PerformLayout

0.57

ysty

0.57

Activations Density 0.174%

commands or suggestions related to user actions

No Comments

No Known Activations