INDEX

Explanations

references to threats and menaces

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

ski

-0.71

fjspx

-0.71

tyfik

-0.66

abin

-0.64

yled

-0.64

 Surname

-0.64

 específicos

-0.62

TypedDataSet

-0.62

lename

-0.61

 مرئيه

-0.61

POSITIVE LOGITS

 threat

2.71

threat

2.63

 Threat

2.61

 threats

2.60

Threat

2.49

 Threats

2.46

 threatened

2.24

 threatening

2.17

Threats

2.14

 threaten

2.12

Activations Density 0.089%

references to threats and menaces

No Comments

No Known Activations