INDEX

Explanations

references to features or characteristics in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

nonatomic

-0.44

 <>",

-0.41

autoconfigure

-0.36

 artific

-0.34

 HttpStatusCode

-0.33

MfG

-0.33

#+#

-0.33

Ett

-0.32

pun

-0.31

 Especially

-0.31

POSITIVE LOGITS

 feature

1.08

 featured

1.06

 pointer

1.00

 Feature

0.87

 pointers

0.86

feature

0.84

pointer

0.83

 FEATURE

0.82

Feature

0.80

ptr

0.77

Activations Density 0.187%

references to features or characteristics in various contexts

No Comments

No Known Activations