INDEX

Explanations

occurrences of racing terminology and events related to pit stops

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

stateProvider

-0.65

 ſtate

-0.60

IUrlHelper

-0.60

__":

-0.57

iſen

-0.56

 AssemblyTitle

-0.54

 purpoſe

-0.53

ſelf

-0.52

 ſch

-0.50

 houſe

-0.50

POSITIVE LOGITS

kissen

0.47

postmedia

0.46

 CAUTION

0.35

amse

0.35

 shift

0.33

graphiques

0.32

 station

0.31

 WARNING

0.31

 timing

0.31

 stint

0.31

Activations Density 0.005%

occurrences of racing terminology and events related to pit stops

No Comments

No Known Activations