INDEX

Explanations

phrases indicating features and functionality of software tools or programming concepts

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

-0.60

WriteBarrier

-0.59

BASELINE

-0.58

存于互联网档案馆

-0.58

 lenker

-0.57

 SqlParameter

-0.57

 nahilalakip

-0.57

createTextNode

-0.56

 ſont

-0.56

 يتيمه

-0.54

POSITIVE LOGITS

欠

0.44

 usage

0.41

 útil

0.36

 useful

0.36

 techniques

0.36

活用

0.36

Usage

0.35

Use

0.35

 using

0.35

Multi

0.35

Activations Density 0.021%

phrases indicating features and functionality of software tools or programming concepts

No Comments

No Known Activations