Hugging Face's logo

Datasets:
ej2
/
transfer

Dataset card Files Community
1
ej2 commited on
Commit
d2b3d3c
·
verified ·
1 Parent(s): 83e7b6b

Upload folder 16b_token

Browse files
Files changed (4) hide show
  1. deepspeed/nozero.json +20 -0
  2. deepspeed/zero1.json +23 -0
  3. deepspeed/zero2.json +23 -0
  4. deepspeed/zero3.json +28 -0
deepspeed/nozero.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "fp16": {
3
+ "enabled": "auto",
4
+ "loss_scale": 0,
5
+ "loss_scale_window": 1000,
6
+ "initial_scale_power": 16,
7
+ "hysteresis": 2,
8
+ "min_loss_scale": 1
9
+ },
10
+ "bf16": {
11
+ "enabled": "auto"
12
+ },
13
+ "train_micro_batch_size_per_gpu": "auto",
14
+ "train_batch_size": "auto",
15
+ "gradient_accumulation_steps": "auto",
16
+ "zero_optimization": {
17
+ "stage": 0
18
+ },
19
+ "gradient_clipping": 6.0
20
+ }
deepspeed/zero1.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "fp16": {
3
+ "enabled": "auto",
4
+ "loss_scale": 0,
5
+ "loss_scale_window": 1000,
6
+ "initial_scale_power": 16,
7
+ "hysteresis": 2,
8
+ "min_loss_scale": 1
9
+ },
10
+ "bf16": {
11
+ "enabled": "auto"
12
+ },
13
+ "train_micro_batch_size_per_gpu": "auto",
14
+ "train_batch_size": "auto",
15
+ "gradient_accumulation_steps": "auto",
16
+ "zero_optimization": {
17
+ "stage": 1,
18
+ "reduce_scatter": true,
19
+ "contiguous_gradients": true,
20
+ "overlap_comm": true,
21
+ "allgather_partitions": true
22
+ }
23
+ }
deepspeed/zero2.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "fp16": {
3
+ "enabled": "auto",
4
+ "loss_scale": 0,
5
+ "loss_scale_window": 1000,
6
+ "initial_scale_power": 16,
7
+ "hysteresis": 2,
8
+ "min_loss_scale": 1
9
+ },
10
+ "bf16": {
11
+ "enabled": "auto"
12
+ },
13
+ "train_micro_batch_size_per_gpu": "auto",
14
+ "train_batch_size": "auto",
15
+ "gradient_accumulation_steps": "auto",
16
+ "zero_optimization": {
17
+ "stage": 2,
18
+ "overlap_comm": true,
19
+ "contiguous_gradients": true,
20
+ "sub_group_size": 1e9,
21
+ "reduce_bucket_size": "auto"
22
+ }
23
+ }
deepspeed/zero3.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "fp16": {
3
+ "enabled": "auto",
4
+ "loss_scale": 0,
5
+ "loss_scale_window": 1000,
6
+ "initial_scale_power": 16,
7
+ "hysteresis": 2,
8
+ "min_loss_scale": 1
9
+ },
10
+ "bf16": {
11
+ "enabled": "auto"
12
+ },
13
+ "train_micro_batch_size_per_gpu": "auto",
14
+ "train_batch_size": "auto",
15
+ "gradient_accumulation_steps": "auto",
16
+ "zero_optimization": {
17
+ "stage": 3,
18
+ "overlap_comm": true,
19
+ "contiguous_gradients": true,
20
+ "sub_group_size": 1e9,
21
+ "reduce_bucket_size": "auto",
22
+ "stage3_prefetch_bucket_size": "auto",
23
+ "stage3_param_persistence_threshold": "auto",
24
+ "stage3_max_live_parameters": 1e9,
25
+ "stage3_max_reuse_distance": 1e9,
26
+ "stage3_gather_16bit_weights_on_model_save": true
27
+ }
28
+ }