Training in progress, step 500

ipu_config.json

@@ -7,7 +7,7 @@
   "execute_encoder_on_cpu_for_generation": false,
   "gradient_accumulation_steps": 16,
   "inference_device_iterations": 5,
-  "inference_replication_factor": 4,
+  "inference_replication_factor": 1,
   "ipus_per_replica": 4,
   "layers_per_ipu": [
     0,
@@ -20,8 +20,8 @@
   "optimum_version": "1.5.1",
   "output_mode": "final",
   "recompute_checkpoint_every_layer": true,
-  "replicated_tensor_sharding": true,
-  "replication_factor": 4,
+  "replicated_tensor_sharding": false,
+  "replication_factor": 1,
   "seed": 42,
   "sharded_execution_for_inference": false,
   "transformers_version": "4.20.1"

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cb0b3b85875e7e579e6cc529986b22366999bcbb5196922e94119abd99aa04e0
+oid sha256:cd7cebe109f17f18ca213809f76282c5d1506ac6c2021df5c6cc89ccffbbad78
 size 217864753

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e5f38e8847151afdc0e89dd45a78b82136ed9a951f749e8c451deb5373398fa8
+oid sha256:941748645259e4a9e04fa517703d9b1d3e701a5d192f13e4b32935a850e9d15a
 size 2735