INDEX

Explanations

references to directed graphical 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 @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

⤹

-0.48

 dAtA

-0.40

 Chwiliwch

-0.40

UrlResolution

-0.36

 насељу

-0.35

brenner

-0.34

 RadioButton

-0.33

 AssertionError

-0.32

ziell

-0.32

 ThemeData

-0.32

POSITIVE LOGITS

RG

1.27

HG

1.25

FG

1.21

TG

1.20

DG

1.20

PG

1.20

SG

1.19

JG

1.18

WG

1.16

KG

1.16

Activations Density 3.354%

references to directed graphical models and their properties

No Comments

No Known Activations