INDEX

Explanations

phrases indicating significance and importance regarding various topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 autorytatywna

-0.78

LookAnd

-0.69

OGND

-0.68

 CreateTagHelper

-0.67

JspWriter

-0.61

 IEnumerator

-0.60

+};

-0.59

StoryboardSegue

-0.59

awtextra

-0.59

äher

-0.59

POSITIVE LOGITS

 important

2.55

important

2.17

 crucial

2.15

 importance

2.11

 vital

2.07

 Important

2.06

Important

1.96

 critical

1.91

重要

1.89

 essential

1.88

Activations Density 0.621%

phrases indicating significance and importance regarding various topics

No Comments

No Known Activations