INDEX

Explanations

terms related to botanical classifications and descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 pinulongan

-0.81

+#+

-0.81

Personensuche

-0.69

nloa

-0.69

twimg

-0.69

 NSCoder

-0.67

 invokingState

-0.66

 Paglinawan

-0.66

ArgsConstructor

-0.66

ngdoc

-0.65

POSITIVE LOGITS

 rigueur

0.49

wohnung

0.49

 Meksiku

0.48

triangleq

0.47

ksessa

0.47

 człowie

0.47

ntomas

0.47

ksesta

0.46

 SqlCommand

0.45

stoffen

0.44

Activations Density 0.028%

terms related to botanical classifications and descriptions

No Comments

No Known Activations