graphcore-rahult
commited on
Commit
•
826cd98
1
Parent(s):
6d83e1f
Training in progress, step 500
Browse files- ipu_config.json +3 -3
- pytorch_model.bin +1 -1
- training_args.bin +1 -1
ipu_config.json
CHANGED
@@ -7,7 +7,7 @@
|
|
7 |
"execute_encoder_on_cpu_for_generation": false,
|
8 |
"gradient_accumulation_steps": 32,
|
9 |
"inference_device_iterations": 5,
|
10 |
-
"inference_replication_factor":
|
11 |
"ipus_per_replica": 4,
|
12 |
"layers_per_ipu": [
|
13 |
0,
|
@@ -22,8 +22,8 @@
|
|
22 |
"optimum_version": "1.5.1",
|
23 |
"output_mode": "final",
|
24 |
"recompute_checkpoint_every_layer": true,
|
25 |
-
"replicated_tensor_sharding":
|
26 |
-
"replication_factor":
|
27 |
"seed": 42,
|
28 |
"sharded_execution_for_inference": false,
|
29 |
"transformers_version": "4.20.1"
|
|
|
7 |
"execute_encoder_on_cpu_for_generation": false,
|
8 |
"gradient_accumulation_steps": 32,
|
9 |
"inference_device_iterations": 5,
|
10 |
+
"inference_replication_factor": 1,
|
11 |
"ipus_per_replica": 4,
|
12 |
"layers_per_ipu": [
|
13 |
0,
|
|
|
22 |
"optimum_version": "1.5.1",
|
23 |
"output_mode": "final",
|
24 |
"recompute_checkpoint_every_layer": true,
|
25 |
+
"replicated_tensor_sharding": false,
|
26 |
+
"replication_factor": 1,
|
27 |
"seed": 42,
|
28 |
"sharded_execution_for_inference": false,
|
29 |
"transformers_version": "4.20.1"
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 248181489
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3d31d8999032ce876523e18d2a975eba339d0963be6c0856c54b789752179c3f
|
3 |
size 248181489
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 2735
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0e59e09e32a5bd2a07f0bf6ed2e257a36d4722870b8bd19642798254c274c9d8
|
3 |
size 2735
|