Training in progress, step 1000

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2697a7f5db8db8c2077dabd8399272a18435138d4f567bd62c16b2cc3caad6af
+oid sha256:17f56497e1b3f1179c4918772a2b10813f081b75ac6f3689acd86d1f779b84ab
 size 483536061

run.sh

@@ -11,13 +11,13 @@ deepspeed run_speech_recognition_seq2seq_streaming-farsipal.py \
 	--torch_compile="True" \
 	--torch_compile_mode="reduce-overhead" \
 	--torch_compile_mode="max-autotune" \
-	--max_steps="10000" \
+	--max_steps="5000" \
 	--output_dir="./" \
 	--per_device_train_batch_size="32" \
 	--gradient_accumulation_steps="1" \
 	--per_device_eval_batch_size="16" \
 	--logging_steps="25" \
-	--learning_rate="3e-9" \
+	--learning_rate="1e-9" \
 	--seed="42" \
 	--warmup_steps="500" \
 	--evaluation_strategy="steps" \

runs/Dec21_06-54-33_129-146-176-120/1671606569.2722359/events.out.tfevents.1671606569.129-146-176-120.2641118.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d304ffafb509f6bed0494593e94782ee8e7ed80e325540be2cdb17c5321f3797
+size 5893

runs/Dec21_06-54-33_129-146-176-120/events.out.tfevents.1671606569.129-146-176-120.2641118.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a9dbe318b2f7f890f338dda80216c1c79f54929795a296c21ae20e85c522cd1
+size 10887

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a592226de0a378161871e811daa0a9fef8ce7c1aacab91f8a63823e0e14b42d9
+oid sha256:937cee168d9a219adaa6bb0c0f3ea3bd771024c3720feb69509f2c45e5c0e369
 size 4731