INDEX

Explanations

references to murder-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 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

 byli

-0.53

 somos

-0.50

 citada

-0.49

↵

-0.49

 Wiktionnaire

-0.48

 recall

-0.48

 smlou

-0.48

 cotidian

-0.48

 acudir

-0.48

 svých

-0.47

POSITIVE LOGITS

########.

1.05

 Савезне

0.91

 Needle

0.88

 needle

0.87

neſs

0.86

needle

0.85

 Мексичка

0.85

Needle

0.78

ineſs

0.77

 فريبيس

0.76

Activations Density 0.095%

references to murder-related terms

No Comments

No Known Activations