INDEX

Explanations

references to historical locations and landmarks

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

DMETHOD

-0.59

tanleria

-0.56

Искәрмәләр

-0.55

 Roskov

-0.54

Spoljašnje

-0.54

 للمعارف

-0.54

Newswire

-0.54

#+#

-0.52

bootstrapcdn

-0.51

seende

-0.51

POSITIVE LOGITS

 Plaque

0.58

 plaque

0.57

 plaques

0.57

 walking

0.56

 CreateTagHelper

0.56

 walks

0.55

MessageOf

0.55

 heritage

0.54

Gizmos

0.54

mne

0.52

Activations Density 0.210%

references to historical locations and landmarks

No Comments

No Known Activations