INDEX

Explanations

references to damage or harm in various contexts

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

 CreateTagHelper

-0.72

 getGroup

-0.67

edoc

-0.66

roxene

-0.64

 betweenstory

-0.61

 referenties

-0.61

Према

-0.59

 Ginn

-0.59

 perek

-0.59

 vouloir

-0.57

POSITIVE LOGITS

 damage

2.35

 Damage

2.29

 DAMAGE

2.21

Damage

2.12

damage

2.11

 damages

2.08

 Damages

2.03

 DAMAGES

1.87

DAMAGE

1.76

 Damaged

1.73

Activations Density 0.074%

references to damage or harm in various contexts

No Comments

No Known Activations