INDEX

Explanations

references to changes 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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

autogui

-0.51

 proficient

-0.47

oton

-0.46

<bos>

-0.45

ot

-0.44

 rudimentary

-0.43

oro

-0.42

 Seelen

-0.42

quirrels

-0.42

topRight

-0.42

POSITIVE LOGITS

change

1.36

 Change

1.35

 change

1.34

Change

1.31

 CHANGE

1.22

CHANGE

1.22

 Changer

1.05

 cambio

1.04

 Changed

1.00

 Changes

0.99

Activations Density 0.063%

references to changes in various contexts

No Comments

No Known Activations