INDEX

Explanations

expressions that imply improvement or enhancement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_11/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

SequentialGroup

-0.72

 LIABILITY

-0.47

skraft

-0.46

igurumi

-0.46

 liability

-0.46

reihe

-0.46

%)$

-0.45

 recevrez

-0.45

 sfera

-0.45

 deafness

-0.45

POSITIVE LOGITS

better

1.41

Better

1.41

 better

1.35

 Better

1.34

 BETTER

1.25

 bessere

0.95

 beter

0.95

 besser

0.92

 mieux

0.90

 besseren

0.87

Activations Density 0.103%

expressions that imply improvement or enhancement

No Comments

No Known Activations