INDEX

Explanations

references to involvement in various contexts or scenarios

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

GEBURTSDATUM

-0.68

abetes

-0.66

 Kebijakan

-0.65

 présidenti

-0.63

 zijne

-0.62

┑

-0.62

 étoit

-0.61

 Geiſt

-0.60

 avoient

-0.60

 mijne

-0.60

POSITIVE LOGITS

 involved

1.24

 involvement

1.14

 Involved

1.13

involved

1.11

 Involvement

1.08

 involve

1.05

 INVOLVED

0.98

 involves

0.96

 involving

0.93

 INVOL

0.89

Activations Density 0.129%

references to involvement in various contexts or scenarios

No Comments

No Known Activations