INDEX

Explanations

geographical or location-related terms and identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_35/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.35.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

 enfans

-1.09

ſelf

-1.04

 pleaſure

-1.02

ſelves

-0.94

RegressionTest

-0.92

 myſelf

-0.91

parsedMessage

-0.90

 itſelf

-0.90

 avoient

-0.85

Efq

-0.84

POSITIVE LOGITS

ng

0.52

đ

0.47

ch

0.46

tr

0.44

cho

0.43

th

0.43

kh

0.43

thu

0.43

sau

0.43

khi

0.42

Activations Density 0.019%

geographical or location-related terms and identifiers

No Comments

No Known Activations