INDEX

Explanations

instances of the word 'aggressive' and other references to assertive behavior

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

 purpoſe

-1.06

 pleaſure

-0.93

 ſta

-0.88

aarrggbb

-0.87

 queſta

-0.86

 faſt

-0.85

ſelf

-0.83

 ſte

-0.78

ſelves

-0.76

 perſon

-0.76

POSITIVE LOGITS

mos

0.70

位

0.65

bit

0.60

MOS

0.58

ir

0.54

bit

0.50

aa

0.49

Mos

0.49

os

0.48

0.47

Activations Density 0.207%

instances of the word 'aggressive' and other references to assertive behavior

No Comments

No Known Activations