INDEX

Explanations

descriptive terms relating to tension or compactness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 Katz

-0.70

trino

-0.62

@"/

-0.61

 Diener

-0.61

 Gerber

-0.59

otoro

-0.59

beelding

-0.58

Ol

-0.58

pel

-0.57

Gem

-0.57

POSITIVE LOGITS

 tight

2.48

 Tight

2.44

Tight

2.39

tight

2.29

 tighter

1.90

 tightness

1.87

 tightly

1.77

 tighten

1.71

 tightened

1.64

 tightening

1.54

Activations Density 0.082%

descriptive terms relating to tension or compactness

No Comments

No Known Activations