INDEX

Explanations

negations and phrases suggesting a lack of presence or participation

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

hatikan

-0.54

 üzere

-0.54

("]");

-0.53

ensivo

-0.53

*}$

-0.51

لاثة

-0.51

 purpoſe

-0.51

 useful

-0.50

Reasonable

-0.50

inafter

-0.49

POSITIVE LOGITS

tagHelperRunner

0.84

InvalidProtocol

0.66

mistak

0.64

 mince

0.59

 spared

0.58

 slowing

0.56

Jeografia

0.55

="@+

0.55

 relâche

0.53

 relent

0.53

Activations Density 0.256%

negations and phrases suggesting a lack of presence or participation

No Comments

No Known Activations