INDEX

Explanations

references to participation or involvement in activities

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 salvare

-0.49

<bos>

-0.49

 Kepala

-0.47

 lhes

-0.46

makeText

-0.45

<!--

-0.45

 préstamos

-0.45

 nukes

-0.45

newFile

-0.44

 namor

-0.43

POSITIVE LOGITS

 participant

1.23

 participate

1.21

 participation

1.19

 Participation

1.14

 Participant

1.12

Participation

1.11

 Participate

1.11

 participated

1.11

 participating

1.10

particip

1.07

Activations Density 0.120%

references to participation or involvement in activities

No Comments

No Known Activations