INDEX

Explanations

references to stages or phases across various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 noite

-0.59

 manhã

-0.57

carbox

-0.56

 hại

-0.53

rophobic

-0.51

IBE

-0.50

terne

-0.49

 Manning

-0.48

axx

-0.48

wrappers

-0.47

POSITIVE LOGITS

 stages

1.49

 stage

1.45

 phases

1.40

 étape

1.39

 Stages

1.37

 phase

1.37

 Stage

1.32

Stage

1.32

stage

1.30

 Phases

1.29

Activations Density 0.263%

references to stages or phases across various contexts

No Comments

No Known Activations