INDEX

Explanations

code-related terms and structures associated with namespaces and using directives in programming

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 17-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_17/width_16k/average_l0_73

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.17.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

<eos>

-0.40

XtraBars

-0.39

 Riggs

-0.37

 remaining

-0.36

iNdEx

-0.35

illoin

-0.35

jubljana

-0.35

 inſ

-0.35

 calificación

-0.34

 extré

-0.33

POSITIVE LOGITS

using

1.44

 AssemblyTitle

1.33

USING

0.98

Using

0.85

namespace

0.84

 using

0.83

__(/*!

0.81

import

0.80

 USING

0.76

 Using

0.75

Activations Density 0.008%

code-related terms and structures associated with namespaces and using directives in programming

No Comments

No Known Activations