INDEX

Explanations

terms related to programming, coding, or technical error messages

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

いれば

-0.43

cování

-0.43

base

-0.42

あれば

-0.42

IL

-0.41

FORMANCE

-0.40

ęb

-0.40

sonder

-0.40

 toppings

-0.39

CommonModule

-0.39

POSITIVE LOGITS

AndEndTag

0.69

 '\\;'

0.66

UnsafeEnabled

0.60

 disambiguazione

0.58

 صوتيه

0.57

 bezeichneter

0.57

日閲覧

0.56

MigrationBuilder

0.56

Dziękuję

0.55

transQ

0.54

Activations Density 0.082%

terms related to programming, coding, or technical error messages

No Comments

No Known Activations