/usr/data/condor/execute/dir_314523/whisper_slu
PeftModel(
  (base_model): LoraModel(
    (model): WhisperSLU(
      (model): WhisperModel(
        (encoder): WhisperEncoder(
          (conv1): Conv1d(80, 768, kernel_size=(3,), stride=(1,), padding=(1,))
          (conv2): Conv1d(768, 768, kernel_size=(3,), stride=(2,), padding=(1,))
          (embed_positions): Embedding(1500, 768)
          (layers): ModuleList(
            (0-11): 12 x WhisperEncoderLayer(
              (self_attn): WhisperAttention(
                (k_proj): Linear(in_features=768, out_features=768, bias=False)
                (v_proj): Linear(in_features=768, out_features=768, bias=True)
                (q_proj): Linear(in_features=768, out_features=768, bias=True)
                (out_proj): Linear(in_features=768, out_features=768, bias=True)
              )
              (self_attn_layer_norm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
              (activation_fn): GELUActivation()
              (fc1): Linear(in_features=768, out_features=3072, bias=True)
              (fc2): Linear(in_features=3072, out_features=768, bias=True)
              (final_layer_norm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
            )
          )
          (layer_norm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
        )
        (decoder): WhisperDecoder(
          (embed_tokens): Embedding(51865, 768, padding_idx=50257)
          (embed_positions): WhisperPositionalEmbedding(448, 768)
          (layers): ModuleList(
            (0-11): 12 x WhisperDecoderLayer(
              (self_attn): WhisperAttention(
                (k_proj): lora.Linear(
                  (base_layer): Linear(in_features=768, out_features=768, bias=False)
                  (lora_dropout): ModuleDict(
                    (default): Dropout(p=0.1, inplace=False)
                  )
                  (lora_A): ModuleDict(
                    (default): Linear(in_features=768, out_features=8, bias=False)
                  )
                  (lora_B): ModuleDict(
                    (default): Linear(in_features=8, out_features=768, bias=False)
                  )
                  (lora_embedding_A): ParameterDict()
                  (lora_embedding_B): ParameterDict()
                )
                (v_proj): lora.Linear(
                  (base_layer): Linear(in_features=768, out_features=768, bias=True)
                  (lora_dropout): ModuleDict(
                    (default): Dropout(p=0.1, inplace=False)
                  )
                  (lora_A): ModuleDict(
                    (default): Linear(in_features=768, out_features=8, bias=False)
                  )
                  (lora_B): ModuleDict(
                    (default): Linear(in_features=8, out_features=768, bias=False)
                  )
                  (lora_embedding_A): ParameterDict()
                  (lora_embedding_B): ParameterDict()
                )
                (q_proj): lora.Linear(
                  (base_layer): Linear(in_features=768, out_features=768, bias=True)
                  (lora_dropout): ModuleDict(
                    (default): Dropout(p=0.1, inplace=False)
                  )
                  (lora_A): ModuleDict(
                    (default): Linear(in_features=768, out_features=8, bias=False)
                  )
                  (lora_B): ModuleDict(
                    (default): Linear(in_features=8, out_features=768, bias=False)
                  )
                  (lora_embedding_A): ParameterDict()
                  (lora_embedding_B): ParameterDict()
                )
                (out_proj): lora.Linear(
                  (base_layer): Linear(in_features=768, out_features=768, bias=True)
                  (lora_dropout): ModuleDict(
                    (default): Dropout(p=0.1, inplace=False)
                  )
                  (lora_A): ModuleDict(
                    (default): Linear(in_features=768, out_features=8, bias=False)
                  )
                  (lora_B): ModuleDict(
                    (default): Linear(in_features=8, out_features=768, bias=False)
                  )
                  (lora_embedding_A): ParameterDict()
                  (lora_embedding_B): ParameterDict()
                )
              )
              (activation_fn): GELUActivation()
              (self_attn_layer_norm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
              (encoder_attn): WhisperAttention(
                (k_proj): lora.Linear(
                  (base_layer): Linear(in_features=768, out_features=768, bias=False)
                  (lora_dropout): ModuleDict(
                    (default): Dropout(p=0.1, inplace=False)
                  )
                  (lora_A): ModuleDict(
                    (default): Linear(in_features=768, out_features=8, bias=False)
                  )
                  (lora_B): ModuleDict(
                    (default): Linear(in_features=8, out_features=768, bias=False)
                  )
                  (lora_embedding_A): ParameterDict()
                  (lora_embedding_B): ParameterDict()
                )
                (v_proj): lora.Linear(
                  (base_layer): Linear(in_features=768, out_features=768, bias=True)
                  (lora_dropout): ModuleDict(
                    (default): Dropout(p=0.1, inplace=False)
                  )
                  (lora_A): ModuleDict(
                    (default): Linear(in_features=768, out_features=8, bias=False)
                  )
                  (lora_B): ModuleDict(
                    (default): Linear(in_features=8, out_features=768, bias=False)
                  )
                  (lora_embedding_A): ParameterDict()
                  (lora_embedding_B): ParameterDict()
                )
                (q_proj): lora.Linear(
                  (base_layer): Linear(in_features=768, out_features=768, bias=True)
                  (lora_dropout): ModuleDict(
                    (default): Dropout(p=0.1, inplace=False)
                  )
                  (lora_A): ModuleDict(
                    (default): Linear(in_features=768, out_features=8, bias=False)
                  )
                  (lora_B): ModuleDict(
                    (default): Linear(in_features=8, out_features=768, bias=False)
                  )
                  (lora_embedding_A): ParameterDict()
                  (lora_embedding_B): ParameterDict()
                )
                (out_proj): lora.Linear(
                  (base_layer): Linear(in_features=768, out_features=768, bias=True)
                  (lora_dropout): ModuleDict(
                    (default): Dropout(p=0.1, inplace=False)
                  )
                  (lora_A): ModuleDict(
                    (default): Linear(in_features=768, out_features=8, bias=False)
                  )
                  (lora_B): ModuleDict(
                    (default): Linear(in_features=8, out_features=768, bias=False)
                  )
                  (lora_embedding_A): ParameterDict()
                  (lora_embedding_B): ParameterDict()
                )
              )
              (encoder_attn_layer_norm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
              (fc1): lora.Linear(
                (base_layer): Linear(in_features=768, out_features=3072, bias=True)
                (lora_dropout): ModuleDict(
                  (default): Dropout(p=0.1, inplace=False)
                )
                (lora_A): ModuleDict(
                  (default): Linear(in_features=768, out_features=8, bias=False)
                )
                (lora_B): ModuleDict(
                  (default): Linear(in_features=8, out_features=3072, bias=False)
                )
                (lora_embedding_A): ParameterDict()
                (lora_embedding_B): ParameterDict()
              )
              (fc2): lora.Linear(
                (base_layer): Linear(in_features=3072, out_features=768, bias=True)
                (lora_dropout): ModuleDict(
                  (default): Dropout(p=0.1, inplace=False)
                )
                (lora_A): ModuleDict(
                  (default): Linear(in_features=3072, out_features=8, bias=False)
                )
                (lora_B): ModuleDict(
                  (default): Linear(in_features=8, out_features=768, bias=False)
                )
                (lora_embedding_A): ParameterDict()
                (lora_embedding_B): ParameterDict()
              )
              (final_layer_norm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
            )
          )
          (layer_norm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
        )
      )
      (proj_out): Linear(in_features=768, out_features=51865, bias=False)
      (classifier): WhisperClassificationHead(
        (embed_positions): WhisperPositionalEmbedding(448, 768)
        (layers): ModuleList(
          (0-1): 2 x WhisperEncoderLayer(
            (self_attn): WhisperAttention(
              (k_proj): lora.Linear(
                (base_layer): Linear(in_features=768, out_features=768, bias=False)
                (lora_dropout): ModuleDict(
                  (default): Dropout(p=0.1, inplace=False)
                )
                (lora_A): ModuleDict(
                  (default): Linear(in_features=768, out_features=8, bias=False)
                )
                (lora_B): ModuleDict(
                  (default): Linear(in_features=8, out_features=768, bias=False)
                )
                (lora_embedding_A): ParameterDict()
                (lora_embedding_B): ParameterDict()
              )
              (v_proj): lora.Linear(
                (base_layer): Linear(in_features=768, out_features=768, bias=True)
                (lora_dropout): ModuleDict(
                  (default): Dropout(p=0.1, inplace=False)
                )
                (lora_A): ModuleDict(
                  (default): Linear(in_features=768, out_features=8, bias=False)
                )
                (lora_B): ModuleDict(
                  (default): Linear(in_features=8, out_features=768, bias=False)
                )
                (lora_embedding_A): ParameterDict()
                (lora_embedding_B): ParameterDict()
              )
              (q_proj): lora.Linear(
                (base_layer): Linear(in_features=768, out_features=768, bias=True)
                (lora_dropout): ModuleDict(
                  (default): Dropout(p=0.1, inplace=False)
                )
                (lora_A): ModuleDict(
                  (default): Linear(in_features=768, out_features=8, bias=False)
                )
                (lora_B): ModuleDict(
                  (default): Linear(in_features=8, out_features=768, bias=False)
                )
                (lora_embedding_A): ParameterDict()
                (lora_embedding_B): ParameterDict()
              )
              (out_proj): lora.Linear(
                (base_layer): Linear(in_features=768, out_features=768, bias=True)
                (lora_dropout): ModuleDict(
                  (default): Dropout(p=0.1, inplace=False)
                )
                (lora_A): ModuleDict(
                  (default): Linear(in_features=768, out_features=8, bias=False)
                )
                (lora_B): ModuleDict(
                  (default): Linear(in_features=8, out_features=768, bias=False)
                )
                (lora_embedding_A): ParameterDict()
                (lora_embedding_B): ParameterDict()
              )
            )
            (self_attn_layer_norm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
            (activation_fn): GELUActivation()
            (fc1): lora.Linear(
              (base_layer): Linear(in_features=768, out_features=2048, bias=True)
              (lora_dropout): ModuleDict(
                (default): Dropout(p=0.1, inplace=False)
              )
              (lora_A): ModuleDict(
                (default): Linear(in_features=768, out_features=8, bias=False)
              )
              (lora_B): ModuleDict(
                (default): Linear(in_features=8, out_features=2048, bias=False)
              )
              (lora_embedding_A): ParameterDict()
              (lora_embedding_B): ParameterDict()
            )
            (fc2): lora.Linear(
              (base_layer): Linear(in_features=2048, out_features=768, bias=True)
              (lora_dropout): ModuleDict(
                (default): Dropout(p=0.1, inplace=False)
              )
              (lora_A): ModuleDict(
                (default): Linear(in_features=2048, out_features=8, bias=False)
              )
              (lora_B): ModuleDict(
                (default): Linear(in_features=8, out_features=768, bias=False)
              )
              (lora_embedding_A): ParameterDict()
              (lora_embedding_B): ParameterDict()
            )
            (final_layer_norm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
          )
        )
        (layer_norm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
        (out_proj): lora.Linear(
          (base_layer): Linear(in_features=768, out_features=37, bias=True)
          (lora_dropout): ModuleDict(
            (default): Dropout(p=0.1, inplace=False)
          )
          (lora_A): ModuleDict(
            (default): Linear(in_features=768, out_features=8, bias=False)
          )
          (lora_B): ModuleDict(
            (default): Linear(in_features=8, out_features=37, bias=False)
          )
          (lora_embedding_A): ParameterDict()
          (lora_embedding_B): ParameterDict()
        )
        (crf): ConditionalRandomField()
      )
    )
  )
)
{'loss': 0.4292, 'learning_rate': 5e-05, 'epoch': 0.18}
{'loss': 0.2746, 'learning_rate': 4.994863481875841e-05, 'epoch': 0.36}
{'eval_loss': 0.26023727655410767, 'eval_f1_score': 0.6564825695260478, 'eval_label_f1': 0.8343125734430082, 'eval_wer': 0.10898676368139949, 'eval_runtime': 344.2386, 'eval_samples_per_second': 2.905, 'eval_steps_per_second': 0.363, 'epoch': 0.36}
{'loss': 0.2568, 'learning_rate': 4.979475034558115e-05, 'epoch': 0.54}
{'loss': 0.2481, 'learning_rate': 4.9538978924776634e-05, 'epoch': 0.71}
{'eval_loss': 0.246540829539299, 'eval_f1_score': 0.6577916992952232, 'eval_label_f1': 0.8347689898198903, 'eval_wer': 0.10217509095131203, 'eval_runtime': 341.9234, 'eval_samples_per_second': 2.925, 'eval_steps_per_second': 0.366, 'epoch': 0.71}
{'loss': 0.2412, 'learning_rate': 4.9182371575975736e-05, 'epoch': 0.89}
{'loss': 0.2385, 'learning_rate': 4.8726393675266716e-05, 'epoch': 1.07}
{'eval_loss': 0.24104812741279602, 'eval_f1_score': 0.6684952978056427, 'eval_label_f1': 0.8322884012539185, 'eval_wer': 0.10484557628299404, 'eval_runtime': 342.1463, 'eval_samples_per_second': 2.923, 'eval_steps_per_second': 0.365, 'epoch': 1.07}
{'loss': 0.2325, 'learning_rate': 4.817291893365055e-05, 'epoch': 1.25}
{'loss': 0.2316, 'learning_rate': 4.752422169756048e-05, 'epoch': 1.43}
{'eval_loss': 0.23740312457084656, 'eval_f1_score': 0.6724477729601892, 'eval_label_f1': 0.8316909735908553, 'eval_wer': 0.10221379363727842, 'eval_runtime': 344.4982, 'eval_samples_per_second': 2.903, 'eval_steps_per_second': 0.363, 'epoch': 1.43}
{'loss': 0.2304, 'learning_rate': 4.678296760308474e-05, 'epoch': 1.61}
{'loss': 0.2291, 'learning_rate': 4.595220262229601e-05, 'epoch': 1.79}
{'eval_loss': 0.2348490208387375, 'eval_f1_score': 0.6698076168040833, 'eval_label_f1': 0.8292108362779742, 'eval_wer': 0.09683412028794798, 'eval_runtime': 338.1818, 'eval_samples_per_second': 2.957, 'eval_steps_per_second': 0.37, 'epoch': 1.79}
{'loss': 0.2275, 'learning_rate': 4.503534054669892e-05, 'epoch': 1.97}
{'loss': 0.2205, 'learning_rate': 4.4036148959228365e-05, 'epoch': 2.14}
{'eval_loss': 0.2333754003047943, 'eval_f1_score': 0.6744822196170379, 'eval_label_f1': 0.8339194998046112, 'eval_wer': 0.09636968805635111, 'eval_runtime': 340.8077, 'eval_samples_per_second': 2.934, 'eval_steps_per_second': 0.367, 'epoch': 2.14}
{'loss': 0.2224, 'learning_rate': 4.2958733752443195e-05, 'epoch': 2.32}
{'loss': 0.2211, 'learning_rate': 4.180752225653292e-05, 'epoch': 2.5}
{'eval_loss': 0.2319139689207077, 'eval_f1_score': 0.672933803368586, 'eval_label_f1': 0.8327457892675283, 'eval_wer': 0.09606006656861986, 'eval_runtime': 341.1059, 'eval_samples_per_second': 2.932, 'eval_steps_per_second': 0.366, 'epoch': 2.5}
{'loss': 0.2196, 'learning_rate': 4.058724504646834e-05, 'epoch': 2.68}
{'loss': 0.2163, 'learning_rate': 3.9302916503054246e-05, 'epoch': 2.86}
{'eval_loss': 0.2304619699716568, 'eval_f1_score': 0.6731669266770671, 'eval_label_f1': 0.829953198127925, 'eval_wer': 0.09807260623887298, 'eval_runtime': 340.688, 'eval_samples_per_second': 2.935, 'eval_steps_per_second': 0.367, 'epoch': 2.86}
{'loss': 0.219, 'learning_rate': 3.7959814207763135e-05, 'epoch': 3.04}
{'loss': 0.2108, 'learning_rate': 3.656345725602089e-05, 'epoch': 3.22}
{'eval_loss': 0.22986993193626404, 'eval_f1_score': 0.6734375, 'eval_label_f1': 0.8328125000000001, 'eval_wer': 0.09544082359315736, 'eval_runtime': 338.4413, 'eval_samples_per_second': 2.955, 'eval_steps_per_second': 0.369, 'epoch': 3.22}
{'loss': 0.2132, 'learning_rate': 3.5119583578059846e-05, 'epoch': 3.4}
{'loss': 0.2104, 'learning_rate': 3.363412636053269e-05, 'epoch': 3.57}
{'eval_loss': 0.22974026203155518, 'eval_f1_score': 0.6792156862745098, 'eval_label_f1': 0.8368627450980393, 'eval_wer': 0.09919498413189876, 'eval_runtime': 343.1743, 'eval_samples_per_second': 2.914, 'eval_steps_per_second': 0.364, 'epoch': 3.57}
{'loss': 0.2128, 'learning_rate': 3.211318966577581e-05, 'epoch': 3.75}
{'loss': 0.2124, 'learning_rate': 3.056302334890786e-05, 'epoch': 3.93}
{'eval_loss': 0.2278670072555542, 'eval_f1_score': 0.6781925343811395, 'eval_label_f1': 0.8345776031434184, 'eval_wer': 0.09447325644399722, 'eval_runtime': 338.3142, 'eval_samples_per_second': 2.956, 'eval_steps_per_second': 0.369, 'epoch': 3.93}
{'loss': 0.2077, 'learning_rate': 2.8989997375834482e-05, 'epoch': 4.11}
{'loss': 0.2027, 'learning_rate': 2.7400575647692046e-05, 'epoch': 4.29}
{'eval_loss': 0.22793905436992645, 'eval_f1_score': 0.6789638932496076, 'eval_label_f1': 0.8335949764521193, 'eval_wer': 0.09443455375803081, 'eval_runtime': 338.197, 'eval_samples_per_second': 2.957, 'eval_steps_per_second': 0.37, 'epoch': 4.29}
{'loss': 0.2077, 'learning_rate': 2.5801289439291388e-05, 'epoch': 4.47}
{'loss': 0.2055, 'learning_rate': 2.419871056070862e-05, 'epoch': 4.65}
{'eval_loss': 0.2275087535381317, 'eval_f1_score': 0.6832347140039449, 'eval_label_f1': 0.8347140039447732, 'eval_wer': 0.09493768867559409, 'eval_runtime': 337.608, 'eval_samples_per_second': 2.962, 'eval_steps_per_second': 0.37, 'epoch': 4.65}
{'loss': 0.2073, 'learning_rate': 2.2599424352307957e-05, 'epoch': 4.83}
{'loss': 0.209, 'learning_rate': 2.1010002624165527e-05, 'epoch': 5.0}
{'eval_loss': 0.22685159742832184, 'eval_f1_score': 0.6821766561514195, 'eval_label_f1': 0.833596214511041, 'eval_wer': 0.09826611966870501, 'eval_runtime': 338.9186, 'eval_samples_per_second': 2.951, 'eval_steps_per_second': 0.369, 'epoch': 5.0}
{'loss': 0.2011, 'learning_rate': 1.9436976651092144e-05, 'epoch': 5.18}
{'loss': 0.2017, 'learning_rate': 1.7886810334224192e-05, 'epoch': 5.36}
{'eval_loss': 0.22723452746868134, 'eval_f1_score': 0.6834645669291338, 'eval_label_f1': 0.8346456692913385, 'eval_wer': 0.09791779549500736, 'eval_runtime': 337.6021, 'eval_samples_per_second': 2.962, 'eval_steps_per_second': 0.37, 'epoch': 5.36}
{'loss': 0.2034, 'learning_rate': 1.6365873639467315e-05, 'epoch': 5.54}
{'loss': 0.2029, 'learning_rate': 1.4880416421940155e-05, 'epoch': 5.72}
{'eval_loss': 0.22658005356788635, 'eval_f1_score': 0.6818718049547778, 'eval_label_f1': 0.8320880849390484, 'eval_wer': 0.09656320148618314, 'eval_runtime': 338.8852, 'eval_samples_per_second': 2.951, 'eval_steps_per_second': 0.369, 'epoch': 5.72}
{'loss': 0.202, 'learning_rate': 1.3436542743979125e-05, 'epoch': 5.9}
{'loss': 0.201, 'learning_rate': 1.2040185792236874e-05, 'epoch': 6.08}
{'eval_loss': 0.22656936943531036, 'eval_f1_score': 0.6800472255017709, 'eval_label_f1': 0.8327430145611965, 'eval_wer': 0.09776298475114173, 'eval_runtime': 337.9074, 'eval_samples_per_second': 2.959, 'eval_steps_per_second': 0.37, 'epoch': 6.08}
{'loss': 0.1979, 'learning_rate': 1.0697083496945765e-05, 'epoch': 6.26}
{'loss': 0.1985, 'learning_rate': 9.412754953531663e-06, 'epoch': 6.43}
{'eval_loss': 0.22672241926193237, 'eval_f1_score': 0.6856465005931198, 'eval_label_f1': 0.8335310399367339, 'eval_wer': 0.0994659029336636, 'eval_runtime': 339.7265, 'eval_samples_per_second': 2.944, 'eval_steps_per_second': 0.368, 'epoch': 6.43}
{'loss': 0.2006, 'learning_rate': 8.192477743467078e-06, 'epoch': 6.61}
{'loss': 0.1996, 'learning_rate': 7.041266247556813e-06, 'epoch': 6.79}
{'eval_loss': 0.226467102766037, 'eval_f1_score': 0.686437327006722, 'eval_label_f1': 0.8343218663503361, 'eval_wer': 0.0988079572722347, 'eval_runtime': 338.3929, 'eval_samples_per_second': 2.955, 'eval_steps_per_second': 0.369, 'epoch': 6.79}
{'loss': 0.2002, 'learning_rate': 5.9638510407716394e-06, 'epoch': 6.97}
{'loss': 0.197, 'learning_rate': 4.9646594533010875e-06, 'epoch': 7.15}
{'eval_loss': 0.2263847291469574, 'eval_f1_score': 0.6843354430379748, 'eval_label_f1': 0.8346518987341772, 'eval_wer': 0.09857574115643626, 'eval_runtime': 338.0877, 'eval_samples_per_second': 2.958, 'eval_steps_per_second': 0.37, 'epoch': 7.15}
{'loss': 0.1983, 'learning_rate': 4.047797377703985e-06, 'epoch': 7.33}
{'loss': 0.1985, 'learning_rate': 3.217032396915265e-06, 'epoch': 7.51}
{'eval_loss': 0.22642947733402252, 'eval_f1_score': 0.6837539432176655, 'eval_label_f1': 0.8351735015772871, 'eval_wer': 0.09849833578450344, 'eval_runtime': 337.2581, 'eval_samples_per_second': 2.965, 'eval_steps_per_second': 0.371, 'epoch': 7.51}
{'loss': 0.1967, 'learning_rate': 2.475778302439524e-06, 'epoch': 7.69}
{'loss': 0.1999, 'learning_rate': 1.827081066349459e-06, 'epoch': 7.86}
{'eval_loss': 0.2263396978378296, 'eval_f1_score': 0.6861429135412555, 'eval_label_f1': 0.8345834978286616, 'eval_wer': 0.09776298475114173, 'eval_runtime': 339.5995, 'eval_samples_per_second': 2.945, 'eval_steps_per_second': 0.368, 'epoch': 7.86}
{'loss': 0.1953, 'learning_rate': 1.273606324733284e-06, 'epoch': 8.04}
{'loss': 0.1963, 'learning_rate': 8.176284240242638e-07, 'epoch': 8.22}
{'eval_loss': 0.22643861174583435, 'eval_f1_score': 0.6864139020537124, 'eval_label_f1': 0.8317535545023697, 'eval_wer': 0.09784039012307454, 'eval_runtime': 339.619, 'eval_samples_per_second': 2.944, 'eval_steps_per_second': 0.368, 'epoch': 8.22}
{'loss': 0.1984, 'learning_rate': 4.6102107522336403e-07, 'epoch': 8.4}
{'loss': 0.1977, 'learning_rate': 2.052496544188487e-07, 'epoch': 8.58}
{'eval_loss': 0.22642208635807037, 'eval_f1_score': 0.6874753062030818, 'eval_label_f1': 0.8328723824575267, 'eval_wer': 0.09791779549500736, 'eval_runtime': 337.3275, 'eval_samples_per_second': 2.964, 'eval_steps_per_second': 0.371, 'epoch': 8.58}
{'loss': 0.1979, 'learning_rate': 5.136518124159162e-08, 'epoch': 8.76}
{'loss': 0.1961, 'learning_rate': 0.0, 'epoch': 8.94}
{'eval_loss': 0.22641009092330933, 'eval_f1_score': 0.6872037914691944, 'eval_label_f1': 0.8325434439178515, 'eval_wer': 0.09799520086694016, 'eval_runtime': 337.666, 'eval_samples_per_second': 2.962, 'eval_steps_per_second': 0.37, 'epoch': 8.94}
{'train_runtime': 50642.5955, 'train_samples_per_second': 12.638, 'train_steps_per_second': 0.099, 'train_loss': 0.21677429428100586, 'epoch': 8.94}
***** train metrics *****
  epoch                    =        8.94
  train_loss               =      0.2168
  train_runtime            = 14:04:02.59
  train_samples_per_second =      12.638
  train_steps_per_second   =       0.099
{'eval_loss': 0.22641009092330933, 'eval_f1_score': 0.6872037914691944, 'eval_label_f1': 0.8325434439178515, 'eval_wer': 0.09799520086694016, 'eval_runtime': 339.3736, 'eval_samples_per_second': 2.947, 'eval_steps_per_second': 0.368, 'epoch': 8.94}
***** eval metrics *****
  epoch                   =       8.94
  eval_f1_score           =     0.6872
  eval_label_f1           =     0.8325
  eval_loss               =     0.2264
  eval_runtime            = 0:05:39.37
  eval_samples            =       1000
  eval_samples_per_second =      2.947
  eval_steps_per_second   =      0.368
  eval_wer                =      0.098