INDEX

Explanations

references to danger and risk

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/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.0.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

ſelves

-0.92

berdayakan

-0.91

 itſelf

-0.90

$_(

-0.87

 ſtand

-0.86

thâu

-0.85

 Houſe

-0.83

ſelf

-0.83

Efq

-0.82

expandindo

-0.81

POSITIVE LOGITS

 danger

1.00

 risk

0.97

danger

0.74

 threat

0.71

risk

0.70

 hazard

0.66

 Risk

0.66

 Danger

0.65

 peligro

0.61

Risk

0.60

Activations Density 0.225%

references to danger and risk

No Comments

No Known Activations