INDEX

Explanations

references to "bots" and associated terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

posedge

-0.61

 useHistory

-0.53

 tiegħ

-0.52

 Archipelago

-0.52

dersfield

-0.52

INSEE

-0.49

ipolar

-0.47

ActivityCompat

-0.47

 secession

-0.47

 dekol

-0.47

POSITIVE LOGITS

MathML

0.66

mw

0.52

MB

0.51

MW

0.50

 screw

0.49

MW

0.49

Uns

0.48

uns

0.46

MB

0.46

Bot

0.45

Activations Density 0.340%

references to "bots" and associated terminology

No Comments

No Known Activations