INDEX

Explanations

specific terms related to convolutional neural network architectures and operations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 typelib

-0.71

WebElementEntity

-0.59

 propOrder

-0.57

ReusableCell

-0.54

AddTagHelper

-0.53

PutMapping

-0.52

 clientId

-0.52

 Bekasi

-0.52

 Bogen

-0.52

ResponseWriter

-0.51

POSITIVE LOGITS

 kernel

0.66

 padding

0.66

kernel

0.64

Kernel

0.61

padding

0.57

 pooling

0.56

 strides

0.56

 stride

0.53

 Conv

0.53

pool

0.53

Activations Density 0.183%

specific terms related to convolutional neural network architectures and operations

No Comments

No Known Activations