INDEX

Explanations

phrases and concepts related to starting and rewriting from the beginning

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.33.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

 Permanent

-0.35

onedDateTime

-0.34

 NotImplemented

-0.34

 honorary

-0.33

 permanent

-0.33

 OMITBAD

-0.32

enumii

-0.32

ỡng

-0.32

Osob

-0.32

 Honorary

-0.32

POSITIVE LOGITS

 scratch

1.48

scratch

1.23

 afresh

1.21

Scratch

1.12

 anew

1.03

 fresh

1.00

 Scratch

0.98

fresh

0.98

 starting

0.90

starting

0.89

Activations Density 0.362%

phrases and concepts related to starting and rewriting from the beginning

No Comments

No Known Activations