INDEX

Explanations

names of people and potentially other identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_10/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.10.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

 betweenstory

-0.61

 Chwiliwch

-0.58

TagMode

-0.54

exitRule

-0.50

kháu

-0.47

principalColumn

-0.45

blurRadius

-0.45

XmlAccessType

-0.41

 study

-0.40

 للمعارف

-0.40

POSITIVE LOGITS

Jim

0.64

Joe

0.57

Jim

0.57

Pete

0.55

Bill

0.53

Dave

0.52

Joe

0.52

Bob

0.51

 Pete

0.51

Mike

0.48

Activations Density 0.374%

names of people and potentially other identifiers

No Comments

No Known Activations