INDEX

Explanations

references to skyscrapers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 19-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_19/width_16k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.19.hook_mlp_out

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

son

-0.75

son

-0.66

Son

-0.55

ViewFeatures

-0.53

sons

-0.52

 sons

-0.50

ons

-0.49

Son

-0.47

ReferenceEquals

-0.46

Per

-0.46

POSITIVE LOGITS

 ſte

0.94

 pleaſure

0.93

 ſche

0.93

 ſtand

0.93

ſy

0.92

 purpoſe

0.91

uſ

0.90

 myſelf

0.90

 ſta

0.90

 itſelf

0.88

Activations Density 0.268%

references to skyscrapers

No Comments

No Known Activations