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:f9567e399ebcb3ca37075b4102465b3d46bf4477a4d4c59f4054d0c0a61f2184
+oid sha256:0261bc9cab87db176350a17b787a8251dbb1b19dc28fc6dd36646b884fe0a47a
 size 261514473

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cb77443e5cec8f2ce4a29a2c83da881a0f2be22d89b5279362de7c3c754e4af4
-size 2671
+oid sha256:1b5c7f9eb43046fb02380fad3bd4997732b77f1575e9e57fd388a656a839a7b9
+size 2735