INDEX

Explanations

terms associated with resolution and effectiveness in problem-solving contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

tak

-0.34

executeQuery

-0.33

Lim

-0.33

 mikro

-0.30

クト

-0.30

地

-0.29

 plancher

-0.29

Ret

-0.29

正

-0.29

是不是

-0.28

POSITIVE LOGITS

parsedMessage

0.82

DockStyle

0.76

IContainer

0.68

MLLoader

0.66

 ویکی‌پدیا

0.65

 OMITBAD

0.65



0.64

 <<<<<<<<<<<<<<

0.62

windowFixed

0.62

VersionUID

0.61

Activations Density 0.558%

terms associated with resolution and effectiveness in problem-solving contexts

No Comments

No Known Activations