INDEX

Explanations

references to first occurrences or milestones related to diverse individuals and their achievements

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

 InputDecoration

-0.58

PhysRevD

-0.57

 sayang

-0.55

ództ

-0.55

 بهترین

-0.53

 dalej

-0.52

 terutama

-0.52

uttavia

-0.51

 dernier

-0.50

admir

-0.49

POSITIVE LOGITS

 ever

1.87

ever

1.38

 EVER

1.37

Ever

1.14

 Ever

1.10

 ooit

1.06

 jemals

1.05

EVER

1.02

 überhaupt

0.93

rungsseite

0.87

Activations Density 0.199%

references to first occurrences or milestones related to diverse individuals and their achievements

No Comments

No Known Activations