INDEX

Explanations

terminology and concepts related to deep learning and computer vision techniques

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

Convolutional Neural Network (CNN) architectures

np_acts-logits-general · gemini-2.0-flash

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.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

jenost

-0.46

 manis

-0.38

 Mati

-0.37

CodeAttribute

-0.36

 granules

-0.36

UserScript

-0.35

StructEnd

-0.35

 provocar

-0.35

spoon

-0.35

Handling

-0.34

POSITIVE LOGITS

 model

0.76

 architecture

0.73

 models

0.72

 networks

0.70

AndEndTag

0.68

 network

0.66

モデル

0.65

 modèle

0.65

model

0.65

 Networks

0.65

Activations Density 0.382%

terminology and concepts related to deep learning and computer vision techniques

Convolutional Neural Network (CNN) architectures

No Comments

No Known Activations