INDEX

Explanations

conditional statements or hypotheses in programming and discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 41-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_41/width_16k/average_l0_113

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.41.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

rungsseite

-0.71

+#+#

-0.57

ValueGenerated

-0.52

 lavora

-0.51

çoivent

-0.50

AddTagHelper

-0.50

 arbeitet

-0.49

 aapt

-0.48

Autoritní

-0.48

 хватает

-0.48

POSITIVE LOGITS

 redund

0.43

 ocasion

0.42

 influ

0.41

 conlle

0.40

 derivation

0.40

 motiv

0.39

 provoc

0.39

 signific

0.39

EndGlobalSection

0.38

 سبب

0.38

Activations Density 0.120%

conditional statements or hypotheses in programming and discussions

No Comments

No Known Activations