INDEX

Explanations

references to archival documents and transcription processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_1/width_16k/average_l0_105

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_mlp_out

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

 متعلقه

-0.90

发表于

-0.77

LookAnd

-0.75

ViewFeatures

-0.64

DECREF

-0.63

 فريبيس

-0.62

 AssemblyTitle

-0.62

PreferredItem

-0.61

tvguidetime

-0.59

 getRule

-0.59

POSITIVE LOGITS

VersionUID

0.49

vington

0.48

Obrázky

0.48

itatem

0.45

diagonal

0.45

itania

0.45

pogon

0.44

source

0.44

frame

0.42

 diagon

0.42

Activations Density 0.465%

references to archival documents and transcription processes

No Comments

No Known Activations