Enable torch autocast for albert xxlarge.

README.md

@@ -13,11 +13,7 @@ This model only contains the `GaudiConfig` file for running the [albert-xxlarge-
 **This model contains no model weights, only a GaudiConfig.**
 
 This enables to specify:
-- `use_habana_mixed_precision`: whether to use Habana Mixed Precision (HMP)
-    - `hmp_opt_level`: optimization level for HMP, see [here](https://docs.habana.ai/en/latest/PyTorch/PyTorch_Mixed_Precision/PT_Mixed_Precision.html#configuration-options) for a detailed explanation
-    - `hmp_bf16_ops`: list of operators that should run in bf16
-    - `hmp_fp32_ops`: list of operators that should run in fp32
-    - `hmp_is_verbose`: verbosity
+- `use_torch_autocast`: whether to use PyTorch's autocast mixed precision
 - `use_fused_adam`: whether to use Habana's custom AdamW implementation
 - `use_fused_clip_norm`: whether to use Habana's fused gradient norm clipping operator
 
@@ -45,4 +41,4 @@ python run_qa.py \
   --throughput_warmup_steps 2
 ```
 
-Check the [documentation](https://huggingface.co/docs/optimum/habana/index) out for more advanced usage and examples. 
+Check the [documentation](https://huggingface.co/docs/optimum/habana/index) out for more advanced usage and examples.

gaudi_config.json

@@ -1,28 +1,5 @@
 {
-  "use_habana_mixed_precision": true,
-  "hmp_is_verbose": false,
+  "use_torch_autocast": true,
   "use_fused_adam": true,
-  "use_fused_clip_norm": true,
-  "hmp_bf16_ops": [
-    "add",
-    "addmm",
-    "bmm",
-    "div",
-    "dropout",
-    "gelu",
-    "iadd",
-    "linear",
-    "layer_norm",
-    "matmul",
-    "mm",
-    "rsub",
-    "softmax",
-    "truediv"
-  ],
-  "hmp_fp32_ops": [
-    "embedding",
-    "nll_loss",
-    "log_softmax",
-    "cross_entropy"
-  ]
+  "use_fused_clip_norm": true
 }