 Nel Nibcord
|
37a75d6b96
Fixing tinygrad model
|
8 月之前 |
|
Nel Nibcord
|
0d3abfca95
Made models save properly
|
8 月之前 |
|
Nel Nibcord
|
9283f6d7bd
Correct loss propagation so we can see the actual loss instead of just the requestor shard's loss
|
8 月之前 |
|
Nel Nibcord
|
9eadee310b
Basic model saving
|
8 月之前 |
|
Nel Nibcord
|
38e368f00b
Fixed up the ops so that batches work
|
8 月之前 |
|
Nel Nibcord
|
dd3d99043b
Working distributed training
|
8 月之前 |
|
Nel Nibcord
|
175ebc1c42
Coordination biz
|
8 月之前 |
|
Nel Nibcord
|
3e869051f6
Okay we should probably await the update
|
8 月之前 |
|
Nel Nibcord
|
75c8650f1f
Naive network-propagated loss implementation on MLX
|
8 月之前 |
|
Nel Nibcord
|
836856824e
WIP: Training works on mlx
|
8 月之前 |
|
Nel Nibcord
|
a6fd7a3430
Generalizing some of the dataset biz while also creating uniform batches
|
8 月之前 |
|
Nel Nibcord
|
f5efbe1b8f
Initial distributed evaluation implementation
|
8 月之前 |
 Alex Cheema
|
8a5d212cfc
test 20
|
8 月之前 |
|
Alex Cheema
|
53edb8508b
test 19
|
8 月之前 |
|
Alex Cheema
|
29d9df04bf
test 18
|
8 月之前 |
|
Alex Cheema
|
4d6af6e6ca
test 17
|
8 月之前 |
|
Alex Cheema
|
8c7c156f57
test 16
|
8 月之前 |
|
Alex Cheema
|
310843487f
test 15
|
8 月之前 |
|
Alex Cheema
|
a4b221d0a0
test 14
|
8 月之前 |
|
Alex Cheema
|
286db875de
test 13
|
8 月之前 |
|
Alex Cheema
|
d714e40f62
test 12
|
8 月之前 |
|
Alex Cheema
|
e78ef75531
test 11
|
8 月之前 |
|
Alex Cheema
|
38eaecf087
test 10
|
8 月之前 |
|
Alex Cheema
|
3cf28f8452
test 9
|
8 月之前 |
|
Alex Cheema
|
9ba8bbdd70
test 8
|
8 月之前 |
|
Alex Cheema
|
af6048e373
test 7
|
8 月之前 |
|
Alex Cheema
|
d93b8e8948
test 6
|
8 月之前 |
|
Alex Cheema
|
b69cb49a46
test 5
|
8 月之前 |
|
Alex Cheema
|
cc74b1f9b3
test 4
|
8 月之前 |
|
Alex Cheema
|
e78a52de5f
test 3
|
8 月之前 |
