INDEX

Explanations

terms related to fire, explosions, and hazardous situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 nadzieję

-0.40

 mijne

-0.39

 dureza

-0.36

 arcilla

-0.34

anInt

-0.34

pannt

-0.34

 impresiones

-0.34

 prestaciones

-0.33

Referencies

-0.33

 kommunen

-0.33

POSITIVE LOGITS

 fire

1.00

 Fire

0.85

 fires

0.84

 flames

0.81

Fire

0.81

 burning

0.79

 flammable

0.78

fire

0.77

 burns

0.77

 flame

0.76

Activations Density 0.205%

terms related to fire, explosions, and hazardous situations

No Comments

No Known Activations