INDEX

Explanations

phrases that express the need for explanations or justifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

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

setVerticalGroup

-0.65

 незавершена

-0.63

 autorytatywna

-0.60

WindowConstants

-0.55

findpost

-0.51

Hentet

-0.51

 ब्रेकडाउन

-0.49

Geografi

-0.49

 otomatig

-0.49

 Chwiliwch

-0.49

POSITIVE LOGITS

 Emoji

0.35

Історія

0.32

 løs

0.31

❋

0.31

 percep

0.31

Cov

0.31

PhysRevLett

0.30

方面的

0.30

Santis

0.30

 emoji

0.30

Activations Density 1.180%

phrases that express the need for explanations or justifications

No Comments

No Known Activations