INDEX

Explanations

elements related to community, support, and networking

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

 Paglinawan

-0.86

ſelf

-0.84

 myſelf

-0.82

writeFieldEnd

-0.81

 Normdatei

-0.81

 Jefus

-0.78

 itſelf

-0.77

Efq

-0.74

 uſed

-0.73

 himſelf

-0.73

POSITIVE LOGITS

 fortunately

0.83

それを

0.83

はその

0.81

Fortunately

0.78

そんな

0.76

为此

0.74

对此

0.74

Luckily

0.73

 luckily

0.73

 Fortunately

0.70

Activations Density 0.359%

elements related to community, support, and networking

No Comments

No Known Activations