Training in progress, step 500

config.json

@@ -33,7 +33,7 @@
     }
   },
   "torch_dtype": "float16",
-  "transformers_version": "4.24.0",
+  "transformers_version": "4.20.1",
   "use_cache": true,
   "vocab_size": 50257
 }

ipu_config.json

@@ -1,8 +1,7 @@
 {
-  "decompose_grad_sum": false,
+  "auto_loss_scaling": false,
   "device_iterations": 2,
   "embedding_serialization_factor": 4,
-  "enable_half_first_order_momentum": true,
   "enable_half_partials": true,
   "executable_cache_dir": "/tmp/exe_cache",
   "execute_encoder_on_cpu_for_generation": false,
@@ -18,14 +17,12 @@
   ],
   "matmul_proportion": 0.25,
   "optimizer_state_offchip": true,
-  "optimum_version": "1.4.1",
+  "optimum_version": "1.5.0",
   "output_mode": "final",
-  "profile_dir": "",
   "recompute_checkpoint_every_layer": true,
   "replicated_tensor_sharding": false,
   "replication_factor": 1,
   "seed": 42,
   "sharded_execution_for_inference": false,
-  "transformers_version": "4.24.0",
-  "use_popdist": false
+  "transformers_version": "4.20.1"
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d71fcb9c47ddfb6219024646bd274085a7a7f0deecdfa818f1714446aec3c511
+oid sha256:d7521d9ee5f97eb46090bbb7ef581594348aeefdc5361b928b32bc44cbe44dc9
 size 261514473

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ebeb70a5ce579e4bd4ab1a34cb6d2334008127ecc95bd989e565eb2cee0e2ff3
-size 2607
+oid sha256:660bc5a8e71f0b96d6825f77549cdaf5b1b305bcfdf608098e1e2d6aada0f569
+size 2671