INDEX

Explanations

words related to emotional states and conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

-0.43

-0.42

assert

-0.39

 deseos

-0.39

-0.38

 bomba

-0.38

 époque

-0.38

 paneles

-0.37

POSITIVE LOGITS

LookAnd

1.19

 utafitiHapana

1.18

SequentialGroup

1.06

 CreateTagHelper

1.01

0.97

 مشين

0.97

 GenerationType

0.96

 kaarangay

0.95

Personensuche

0.95

 mergeFrom

0.93

Activations Density 0.102%

words related to emotional states and conditions

No Comments

No Known Activations