INDEX

Explanations

phrases indicating freedom and focus in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 Engle

-0.35

cov

-0.34

vy

-0.33

 academy

-0.33

 entered

-0.33

 Academy

-0.33

 entering

-0.31

academy

-0.29

st

-0.29

iz

-0.29

POSITIVE LOGITS

InitVars

0.73

 ProtoMessage

0.69

awtextra

0.67

 EconPapers

0.65

GEBURTSDATUM

0.64

<unused21>

0.63

<unused23>

0.63

<unused8>

0.63

<unused1>

0.63

<unused3>

0.63

Activations Density 0.182%

phrases indicating freedom and focus in various contexts

No Comments

No Known Activations