INDEX

Explanations

terms related to safety and protection, particularly in the context of guidelines or systems

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

istible

-0.61

principalTable

-0.60

TacToe

-0.58

FORME

-0.57

ביוגרפיה

-0.56

 chré

-0.56

GOTREF

-0.54

netti

-0.54

ξει

-0.54

 jspb

-0.53

POSITIVE LOGITS

 guards

0.96

 guard

0.93

guard

0.89

guards

0.85

Guard

0.76

 Guards

0.73

 Guard

0.70

GUARD

0.68

 GUARD

0.66

 conservation

0.65

Activations Density 0.125%

terms related to safety and protection, particularly in the context of guidelines or systems

No Comments

No Known Activations