Initial rough port of llm.c

[nietras]

llm.cs test.

TENSOR OK
step 0: loss 5.26989 (took 562062.1535 ms)
step 1: loss 4.059388 (took 233129.9491 ms)
step 2: loss 3.374212 (took 213911.9533 ms)
step 3: loss 2.8001275 (took 212057.465 ms)
step 4: loss 2.3153117 (took 212057.61430000002 ms)
step 5: loss 1.8493471 (took 211775.94110000003 ms)
step 6: loss 1.3952174 (took 212021.9995 ms)
step 7: loss 0.9986158 (took 212078.4283 ms)
step 8: loss 0.6255399 (took 212064.9952 ms)
step 9: loss 0.37801182 (took 1286941.2931 ms)
loss ok at step 0: 5.26989 5.270007
loss ok at step 1: 4.059388 4.0597067
loss ok at step 2: 3.374212 3.375123
loss ok at step 3: 2.8001275 2.8007827
loss ok at step 4: 2.3153117 2.3153822
loss ok at step 5: 1.8493471 1.8490286
loss ok at step 6: 1.3952174 1.3946564
loss ok at step 7: 0.9986158 0.9991465
loss ok at step 8: 0.6255399 0.6240804
loss ok at step 9: 0.37801182 0.37651098
overall okay: True

C:\git\oss\llm.cs\artifacts\bin\Llm\debug_net8.0\Llm.exe (process 56636) exited with code 0.

llm.c

TENSOR OK
step 0: loss 5.269890 (took 399782.000000 ms)
step 1: loss 4.059388 (took 44282.000000 ms)
step 2: loss 3.374212 (took 44781.000000 ms)
step 3: loss 2.800128 (took 44687.000000 ms)
step 4: loss 2.315312 (took 44406.000000 ms)
step 5: loss 1.849347 (took 45203.000000 ms)
step 6: loss 1.395217 (took 44719.000000 ms)
step 7: loss 0.998616 (took 44968.000000 ms)
step 8: loss 0.625540 (took 44390.000000 ms)
step 9: loss 0.378012 (took 44172.000000 ms)
loss ok at step 0: 5.269890 5.270007
loss ok at step 1: 4.059388 4.059707
loss ok at step 2: 3.374212 3.375123
loss ok at step 3: 2.800128 2.800783
loss ok at step 4: 2.315312 2.315382
loss ok at step 5: 1.849347 1.849029
loss ok at step 6: 1.395217 1.394656
loss ok at step 7: 0.998616 0.999147
loss ok at step 8: 0.625540 0.624080
loss ok at step 9: 0.378012 0.376511
overall okay: 1

C:\git\oss\llm.c\out\build\x64-Debug\bin\test_gpt2.exe (process 60596) exited with code 0.