INDEX

Explanations

fragments of code or programming syntax

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_0/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.hook_mlp_out

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

deo

-0.57

Примітки

-0.53

inali

-0.52

yn

-0.51

 poils

-0.50

subsubsection

-0.50

交

-0.50

‗

-0.49

تها

-0.49

uker

-0.48

POSITIVE LOGITS

 NSCoder

0.83

 GenerationType

0.79

 MainAxisSize

0.72

rzost

0.66

himovic

0.65

'},

0.64

setVerticalGroup

0.63

 consultato

0.63

bibfield

0.63

abestanden

0.61

Activations Density 0.001%

fragments of code or programming syntax

No Comments

No Known Activations