Search: 152334h.github.io | Heykuki News

HK

Heykuki News

Top New Best Ask Show Jobs

Top New Best Ask Show Jobs

1.

Non-determinism in GPT-4 is caused by Sparse MoE

152334h.github.io

3 years ago

397 points

2.

Calculating the cost of a Google DeepMind paper

152334h.github.io

2 years ago

303 points

3.

Knowing Enough About MoE to Explain Dropped Tokens in GPT-4

152334h.github.io

3 years ago

3 points

4.

Can AI agents write kernel exploits?

152334h.github.io

4 months ago

3 points

5.

Why can't TorToiSe be fine-tuned?

152334h.github.io

3 years ago

1 points