INDEX

Explanations

complex mathematical or programming expressions and notations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

parsedMessage

-1.23

 queſta

-1.20

transQ

-1.13

GraphicsUnit

-1.12

GEBURTSDATUM

-1.05

RegressionTest

-1.04

 autorytatywna

-1.02

<unused41>

-1.00

<unused8>

-1.00

[@BOS@]

-1.00

POSITIVE LOGITS

0.54

0.53

0.50

0.49

0.48

0.44

Activations Density 0.001%

complex mathematical or programming expressions and notations

No Comments

No Known Activations