INDEX

Explanations

technical terms and identifiers related to coding or programming

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.17.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

TabIndex

-0.46

 Hello

-0.45

Hello

-0.44

ดย

-0.43

 more

-0.42

 pulsa

-0.41

 CreateTagHelper

-0.40

drm

-0.40

-0.39

uras

-0.39

POSITIVE LOGITS

 OFDb

0.69

';

0.68

hdysval

0.67

barung

0.66

Pautan

0.66

PRWEB

0.66

%</

0.65

пасибо

0.65

Geografía

0.65

...');

0.64

Activations Density 0.108%

technical terms and identifiers related to coding or programming

No Comments

No Known Activations