INDEX

Explanations

contexts and discussions pertaining to damage, particularly relating to injuries or harm caused in various scenarios

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_17/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.17.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

EndContext

-0.53

 Entire

-0.53

 entire

-0.51

cob

-0.50

 viață

-0.48

Према

-0.47

entire

-0.47

følgelig

-0.46

 Soko

-0.46

ريكا

-0.46

POSITIVE LOGITS

 damage

1.11

 DAMAGE

1.09

 Damage

1.06

damage

1.02

Damage

0.99

 inflicted

0.93

 damages

0.89

 Damaged

0.86

 injure

0.86

 injuries

0.84

Activations Density 0.215%

contexts and discussions pertaining to damage, particularly relating to injuries or harm caused in various scenarios

No Comments

No Known Activations