INDEX

Explanations

references to detention or imprisonment conditions

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 19-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_19/width_16k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.19.hook_mlp_out

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 متعلقه

-0.59

oum

-0.49

roek

-0.49

Morfologia

-0.47

 prodigal

-0.47

rendon

-0.47

hoebe

-0.47

InMillis

-0.47

 pageable

-0.47

ExecuteReader

-0.46

POSITIVE LOGITS

 Nazi

0.83

 Gesta

0.78

 Hitler

0.72

Hitler

0.71

 Axis

0.69

 Nazis

0.68

TestingModule

0.65

NAZ

0.65

 Vichy

0.64

NSD

0.64

Activations Density 0.483%

references to detention or imprisonment conditions

No Known Activations