INDEX

Explanations

references to reports, evaluations, and structured reviews within the document

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 NSCoder

-0.39

 дописавши

-0.38

 vast

-0.36

 ſmall

-0.36

 duquel

-0.35

 fuper

-0.34

positivos

-0.34

 belangrij

-0.33

chartInstance

-0.33

 мәкал

-0.33

POSITIVE LOGITS

 AssemblyTitle

0.65

RegressionTest

0.61

 surla

0.59

 ujednoznacz

0.58

Espèce

0.57

UnitTesting

0.56

UserScript

0.53

transQ

0.52

rungsseite

0.51

arschu

0.51

Activations Density 0.614%

references to reports, evaluations, and structured reviews within the document

No Comments

No Known Activations