INDEX

Explanations

mentions of 'office' and 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 @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

жидан

-0.73

 AssemblyProduct

-0.72

Đây

-0.72

Tembelea

-0.71

toggler

-0.70

denly

-0.70

Cordialement

-0.69

TagMode

-0.69

 تضيفلها

-0.69

theless

-0.68

POSITIVE LOGITS

 office

3.35

 Office

3.10

office

3.01

Office

2.91

 offices

2.91

 OFFICE

2.86

 Offices

2.65

OFFICE

2.58

offices

2.45

 OFFICES

2.12

Activations Density 0.044%

mentions of 'office' and related terms

No Comments

No Known Activations