The outputs of the tf.dot operation on theWasm backend are inconsistent with the output on the TensorFlow backend

[liliquan0118]

Please make sure that this is a bug. As per our
GitHub Policy,
we only address code/doc bugs, performance issues, feature requests and
build/installation issues on GitHub. tag:bug_template

System information

• Have I written custom code (as opposed to using a stock example script provided in TensorFlow.js):
• OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Ubuntu 20.04, Macos 13.3 (22E252)
• Mobile device (e.g. iPhone 8, Pixel 2, Samsung Galaxy) if the issue happens on mobile device:
• TensorFlow.js installed from (npm or script link): script link
• TensorFlow.js version (use command below): 4.2.0
• Browser version: Chrome Version 128.0.6613.86 (Official Build) (arm64)
• Tensorflow.js Converter Version:

Describe the current behavior

Given the "t1": [[1289328924.64505,2051092337.6925373,173654905.43031645,2100669721.3770614],[701451708.9475408,-1034240466.2648284,-1554255981.663638,-1773384464.5712419]], "t2": [[-1434384320.1547623,-595326783.14995],[382591937.3909383,1406908466.487136],[1124709131.724626,-244281148.61611748],[1217546997.2704406,-1925705365.3955224]], tf.dot(t1,t2) return [[1688313573467488300 , -1969564662122938400], [-5309108442948960000, 1922016968974532600 ]] on the Wasm backend. However, tf.dot(t1,t2) return [[1688313573467488300 , -1969564524683985000], [-5309108442948960000, 1922016968974532600 ]] on the tensorflow backend.

Output:

Describe the expected behavior

Standalone code to reproduce the issue
Provide a reproducible test case that is the bare minimum necessary to generate
the problem. If possible, please share a link to Colab/CodePen/any notebook.

async function dot(backend){
    await tf.setBackend(backend);
    await tf.ready() 
    t1 = [[1289328924.64505,2051092337.6925373,173654905.43031645,2100669721.3770614],[701451708.9475408,-1034240466.2648284,-1554255981.663638,-1773384464.5712419]]
    t2 = [[-1434384320.1547623,-595326783.14995],[382591937.3909383,1406908466.487136],[1124709131.724626,-244281148.61611748],[1217546997.2704406,-1925705365.3955224]]
    const result = await tf.dot(t1,t2);
    await console.log("the result of ", tf.getBackend(), "is:\n" );
    await result.print();
}
async function test() {
    await dot("cpu");
    await dot("tensorflow");
    // await dot("wasm");
}

test();

Other info / logs Include any logs or source code that would be helpful to
diagnose the problem. If including tracebacks, please include the full
traceback. Large logs and files should be attached.