INDEX

Explanations

references to people involved in a narrative or legal testimony

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

death and injury

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.24.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

 للاسماء

-0.51

 pravi

-0.42

 Numerade

-0.42

Amicalement

-0.42

galkan

-0.41

WriteBarrier

-0.40

บ้าง

-0.40

AddTagHelper

-0.40

 initializer

-0.39

 vacuna

-0.39

POSITIVE LOGITS

 fatal

1.08

 fatally

1.02

 tragically

0.97

 tragic

0.95

 fatality

0.86

 Fatal

0.83

Trag

0.82

 died

0.81

fatal

0.79

 killed

0.78

Activations Density 0.218%

references to people involved in a narrative or legal testimony

death and injury

No Comments

No Known Activations