INDEX

Explanations

mentions of inappropriate or suggestive content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/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.21.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

 saites

-0.50

tagHelperRunner

-0.49

NamedQueries

-0.47

 propOrder

-0.45

 esternos

-0.44

 mergeFrom

-0.43

-0.42

Atsauces

-0.42

UniformLocation

-0.40

calientes

-0.40

POSITIVE LOGITS

posedge

0.50

 minhas

0.46

 !”

0.45

 miei

0.44

0.42

⋙

0.42

AddTagHelper

0.41

my

0.40

 مشين

0.40

!”

0.40

Activations Density 0.194%

mentions of inappropriate or suggestive content

No Comments

No Known Activations