Training in progress, epoch 1

config.json

@@ -35,5 +35,5 @@
   "torch_dtype": "float32",
   "transformers_version": "4.45.0.dev0",
   "use_cache": true,
-  "vocab_size": 64005
+  "vocab_size": 64006
 }

egy_training_log.txt

@@ -70,311 +70,7 @@ local_rank=0,
 log_level=passive,
 log_level_replica=warning,
 log_on_each_node=True,
-logging_dir=/home/iais_marenpielka/Bouthaina/res_nw_dj/runs/Aug30_21-28-25_lmgpu-node-02,
-logging_first_step=False,
-logging_nan_inf_filter=True,
-logging_steps=500,
-logging_strategy=IntervalStrategy.EPOCH,
-lr_scheduler_kwargs={},
-lr_scheduler_type=SchedulerType.LINEAR,
-max_grad_norm=1.0,
-max_steps=-1,
-metric_for_best_model=loss,
-mp_parameters=,
-neftune_noise_alpha=None,
-no_cuda=False,
-num_train_epochs=3.0,
-optim=OptimizerNames.ADAMW_TORCH,
-optim_args=None,
-optim_target_modules=None,
-output_dir=/home/iais_marenpielka/Bouthaina/res_nw_dj,
-overwrite_output_dir=False,
-past_index=-1,
-per_device_eval_batch_size=8,
-per_device_train_batch_size=8,
-prediction_loss_only=False,
-push_to_hub=True,
-push_to_hub_model_id=None,
-push_to_hub_organization=None,
-push_to_hub_token=<PUSH_TO_HUB_TOKEN>,
-ray_scope=last,
-remove_unused_columns=True,
-report_to=[],
-restore_callback_states_from_checkpoint=False,
-resume_from_checkpoint=None,
-run_name=/home/iais_marenpielka/Bouthaina/res_nw_dj,
-save_on_each_node=False,
-save_only_model=False,
-save_safetensors=True,
-save_steps=500,
-save_strategy=IntervalStrategy.EPOCH,
-save_total_limit=None,
-seed=42,
-skip_memory_metrics=True,
-split_batches=None,
-tf32=None,
-torch_compile=False,
-torch_compile_backend=None,
-torch_compile_mode=None,
-torch_empty_cache_steps=None,
-torchdynamo=None,
-tpu_metrics_debug=False,
-tpu_num_cores=None,
-use_cpu=False,
-use_ipex=False,
-use_legacy_prediction_loop=False,
-use_mps_device=False,
-warmup_ratio=0.0,
-warmup_steps=500,
-weight_decay=0.0,
-)
-INFO:datasets.builder:Using custom data configuration default-2166f9793d70674b
-INFO:datasets.info:Loading Dataset Infos from /home/iais_marenpielka/Bouthaina/miniconda3/lib/python3.12/site-packages/datasets/packaged_modules/text
-INFO:datasets.builder:Generating dataset text (/home/iais_marenpielka/.cache/huggingface/datasets/text/default-2166f9793d70674b/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101)
-INFO:datasets.builder:Downloading and preparing dataset text/default to /home/iais_marenpielka/.cache/huggingface/datasets/text/default-2166f9793d70674b/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101...
-INFO:datasets.download.download_manager:Downloading took 0.0 min
-INFO:datasets.download.download_manager:Checksum Computation took 0.0 min
-INFO:datasets.builder:Generating train split
-INFO:datasets.builder:Generating validation split
-INFO:datasets.utils.info_utils:Unable to verify splits sizes.
-INFO:datasets.builder:Dataset text downloaded and prepared to /home/iais_marenpielka/.cache/huggingface/datasets/text/default-2166f9793d70674b/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101. Subsequent calls will reuse this data.
-INFO:datasets.arrow_dataset:Caching processed dataset at /home/iais_marenpielka/.cache/huggingface/datasets/text/default-2166f9793d70674b/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101/cache-026bde2ce1e4f1eb.arrow
-INFO:datasets.arrow_dataset:Caching processed dataset at /home/iais_marenpielka/.cache/huggingface/datasets/text/default-2166f9793d70674b/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101/cache-f24696270c2f70a2.arrow
-WARNING:__main__:The tokenizer picked seems to have a very large `model_max_length` (1000000000000000019884624838656). Using block_size=768 instead. You can change that default value by passing --block_size xxx.
-INFO:datasets.arrow_dataset:Caching processed dataset at /home/iais_marenpielka/.cache/huggingface/datasets/text/default-2166f9793d70674b/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101/cache-4ebbc72089ee9723.arrow
-INFO:datasets.arrow_dataset:Caching processed dataset at /home/iais_marenpielka/.cache/huggingface/datasets/text/default-2166f9793d70674b/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101/cache-5e8574b46c8155fd.arrow
-WARNING:accelerate.utils.other:Detected kernel version 5.4.0, which is below the recommended minimum of 5.5.0; this can cause the process to hang. It is recommended to upgrade the kernel to the minimum version or higher.
-INFO:root:Epoch 1.0: Train Loss = None, Eval Loss = None
-INFO:absl:Using default tokenizer.
-INFO:root:Epoch 2.0: Train Loss = 1.2513, Eval Loss = 0.7111806869506836
-INFO:absl:Using default tokenizer.
-INFO:root:Epoch 3.0: Train Loss = 0.6462, Eval Loss = 0.6569304466247559
-INFO:absl:Using default tokenizer.
-INFO:__main__:*** Evaluate ***
-INFO:absl:Using default tokenizer.
-WARNING:__main__:Process rank: 0, device: cuda:0, n_gpu: 1, distributed training: False, 16-bits training: False
-INFO:__main__:Training/evaluation parameters TrainingArguments(
-_n_gpu=1,
-accelerator_config={'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False},
-adafactor=False,
-adam_beta1=0.9,
-adam_beta2=0.999,
-adam_epsilon=1e-08,
-auto_find_batch_size=False,
-batch_eval_metrics=False,
-bf16=False,
-bf16_full_eval=False,
-data_seed=None,
-dataloader_drop_last=False,
-dataloader_num_workers=0,
-dataloader_persistent_workers=False,
-dataloader_pin_memory=True,
-dataloader_prefetch_factor=None,
-ddp_backend=None,
-ddp_broadcast_buffers=None,
-ddp_bucket_cap_mb=None,
-ddp_find_unused_parameters=None,
-ddp_timeout=1800,
-debug=[],
-deepspeed=None,
-disable_tqdm=False,
-dispatch_batches=None,
-do_eval=True,
-do_predict=False,
-do_train=True,
-eval_accumulation_steps=None,
-eval_delay=0,
-eval_do_concat_batches=True,
-eval_on_start=False,
-eval_steps=None,
-eval_strategy=IntervalStrategy.EPOCH,
-eval_use_gather_object=False,
-evaluation_strategy=epoch,
-fp16=False,
-fp16_backend=auto,
-fp16_full_eval=False,
-fp16_opt_level=O1,
-fsdp=[],
-fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False},
-fsdp_min_num_params=0,
-fsdp_transformer_layer_cls_to_wrap=None,
-full_determinism=False,
-gradient_accumulation_steps=1,
-gradient_checkpointing=False,
-gradient_checkpointing_kwargs=None,
-greater_is_better=False,
-group_by_length=False,
-half_precision_backend=auto,
-hub_always_push=False,
-hub_model_id=None,
-hub_private_repo=False,
-hub_strategy=HubStrategy.EVERY_SAVE,
-hub_token=<HUB_TOKEN>,
-ignore_data_skip=False,
-include_inputs_for_metrics=False,
-include_num_input_tokens_seen=False,
-include_tokens_per_second=False,
-jit_mode_eval=False,
-label_names=None,
-label_smoothing_factor=0.0,
-learning_rate=5e-05,
-length_column_name=length,
-load_best_model_at_end=True,
-local_rank=0,
-log_level=passive,
-log_level_replica=warning,
-log_on_each_node=True,
-logging_dir=/home/iais_marenpielka/Bouthaina/res_nw_dj/runs/Aug31_10-25-38_lmgpu-node-02,
-logging_first_step=False,
-logging_nan_inf_filter=True,
-logging_steps=500,
-logging_strategy=IntervalStrategy.EPOCH,
-lr_scheduler_kwargs={},
-lr_scheduler_type=SchedulerType.LINEAR,
-max_grad_norm=1.0,
-max_steps=-1,
-metric_for_best_model=loss,
-mp_parameters=,
-neftune_noise_alpha=None,
-no_cuda=False,
-num_train_epochs=3.0,
-optim=OptimizerNames.ADAMW_TORCH,
-optim_args=None,
-optim_target_modules=None,
-output_dir=/home/iais_marenpielka/Bouthaina/res_nw_dj,
-overwrite_output_dir=False,
-past_index=-1,
-per_device_eval_batch_size=8,
-per_device_train_batch_size=8,
-prediction_loss_only=False,
-push_to_hub=True,
-push_to_hub_model_id=None,
-push_to_hub_organization=None,
-push_to_hub_token=<PUSH_TO_HUB_TOKEN>,
-ray_scope=last,
-remove_unused_columns=True,
-report_to=[],
-restore_callback_states_from_checkpoint=False,
-resume_from_checkpoint=None,
-run_name=/home/iais_marenpielka/Bouthaina/res_nw_dj,
-save_on_each_node=False,
-save_only_model=False,
-save_safetensors=True,
-save_steps=500,
-save_strategy=IntervalStrategy.EPOCH,
-save_total_limit=None,
-seed=42,
-skip_memory_metrics=True,
-split_batches=None,
-tf32=None,
-torch_compile=False,
-torch_compile_backend=None,
-torch_compile_mode=None,
-torch_empty_cache_steps=None,
-torchdynamo=None,
-tpu_metrics_debug=False,
-tpu_num_cores=None,
-use_cpu=False,
-use_ipex=False,
-use_legacy_prediction_loop=False,
-use_mps_device=False,
-warmup_ratio=0.0,
-warmup_steps=500,
-weight_decay=0.0,
-)
-INFO:__main__:Checkpoint detected, resuming training at /home/iais_marenpielka/Bouthaina/res_nw_dj/checkpoint-8109. To avoid this behavior, change the `--output_dir` or add `--overwrite_output_dir` to train from scratch.
-INFO:datasets.builder:Using custom data configuration default-98487e126fdb56c7
-INFO:datasets.info:Loading Dataset Infos from /home/iais_marenpielka/Bouthaina/miniconda3/lib/python3.12/site-packages/datasets/packaged_modules/text
-INFO:datasets.builder:Generating dataset text (/home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101)
-INFO:datasets.builder:Downloading and preparing dataset text/default to /home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101...
-INFO:datasets.download.download_manager:Downloading took 0.0 min
-INFO:datasets.download.download_manager:Checksum Computation took 0.0 min
-INFO:datasets.builder:Generating train split
-INFO:datasets.builder:Generating validation split
-INFO:datasets.utils.info_utils:Unable to verify splits sizes.
-INFO:datasets.builder:Dataset text downloaded and prepared to /home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101. Subsequent calls will reuse this data.
-INFO:datasets.arrow_dataset:Caching processed dataset at /home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101/cache-22cb64ac8b531f65.arrow
-INFO:datasets.arrow_dataset:Caching processed dataset at /home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101/cache-7840ce8488671e31.arrow
-WARNING:__main__:The tokenizer picked seems to have a very large `model_max_length` (1000000000000000019884624838656). Using block_size=768 instead. You can change that default value by passing --block_size xxx.
-INFO:datasets.arrow_dataset:Caching processed dataset at /home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101/cache-1f7e65ff1cef4012.arrow
-INFO:datasets.arrow_dataset:Caching processed dataset at /home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101/cache-cb43c016a75212ff.arrow
-WARNING:accelerate.utils.other:Detected kernel version 5.4.0, which is below the recommended minimum of 5.5.0; this can cause the process to hang. It is recommended to upgrade the kernel to the minimum version or higher.
-INFO:__main__:*** Evaluate ***
-INFO:absl:Using default tokenizer.
-WARNING:root:No losses were recorded, so the loss graph was not generated.
-WARNING:__main__:Process rank: 0, device: cuda:0, n_gpu: 1, distributed training: False, 16-bits training: False
-INFO:__main__:Training/evaluation parameters TrainingArguments(
-_n_gpu=1,
-accelerator_config={'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False},
-adafactor=False,
-adam_beta1=0.9,
-adam_beta2=0.999,
-adam_epsilon=1e-08,
-auto_find_batch_size=False,
-batch_eval_metrics=False,
-bf16=False,
-bf16_full_eval=False,
-data_seed=None,
-dataloader_drop_last=False,
-dataloader_num_workers=0,
-dataloader_persistent_workers=False,
-dataloader_pin_memory=True,
-dataloader_prefetch_factor=None,
-ddp_backend=None,
-ddp_broadcast_buffers=None,
-ddp_bucket_cap_mb=None,
-ddp_find_unused_parameters=None,
-ddp_timeout=1800,
-debug=[],
-deepspeed=None,
-disable_tqdm=False,
-dispatch_batches=None,
-do_eval=True,
-do_predict=False,
-do_train=True,
-eval_accumulation_steps=None,
-eval_delay=0,
-eval_do_concat_batches=True,
-eval_on_start=False,
-eval_steps=None,
-eval_strategy=IntervalStrategy.EPOCH,
-eval_use_gather_object=False,
-evaluation_strategy=epoch,
-fp16=False,
-fp16_backend=auto,
-fp16_full_eval=False,
-fp16_opt_level=O1,
-fsdp=[],
-fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False},
-fsdp_min_num_params=0,
-fsdp_transformer_layer_cls_to_wrap=None,
-full_determinism=False,
-gradient_accumulation_steps=1,
-gradient_checkpointing=False,
-gradient_checkpointing_kwargs=None,
-greater_is_better=False,
-group_by_length=False,
-half_precision_backend=auto,
-hub_always_push=False,
-hub_model_id=None,
-hub_private_repo=False,
-hub_strategy=HubStrategy.EVERY_SAVE,
-hub_token=<HUB_TOKEN>,
-ignore_data_skip=False,
-include_inputs_for_metrics=False,
-include_num_input_tokens_seen=False,
-include_tokens_per_second=False,
-jit_mode_eval=False,
-label_names=None,
-label_smoothing_factor=0.0,
-learning_rate=5e-05,
-length_column_name=length,
-load_best_model_at_end=True,
-local_rank=0,
-log_level=passive,
-log_level_replica=warning,
-log_on_each_node=True,
-logging_dir=/home/iais_marenpielka/Bouthaina/res_nw_dj/runs/Aug31_11-16-58_lmgpu-node-02,
+logging_dir=/home/iais_marenpielka/Bouthaina/res_nw_dj/runs/Sep01_13-56-27_lmgpu-node-07,
 logging_first_step=False,
 logging_nan_inf_filter=True,
 logging_steps=500,
@@ -432,187 +128,17 @@ warmup_ratio=0.0,
 warmup_steps=500,
 weight_decay=0.0,
 )
-INFO:__main__:Checkpoint detected, resuming training at /home/iais_marenpielka/Bouthaina/res_nw_dj/checkpoint-8109. To avoid this behavior, change the `--output_dir` or add `--overwrite_output_dir` to train from scratch.
-INFO:datasets.builder:Using custom data configuration default-98487e126fdb56c7
+INFO:datasets.builder:Using custom data configuration default-70891baac37034df
 INFO:datasets.info:Loading Dataset Infos from /home/iais_marenpielka/Bouthaina/miniconda3/lib/python3.12/site-packages/datasets/packaged_modules/text
 INFO:datasets.builder:Overwrite dataset info from restored data version if exists.
-INFO:datasets.info:Loading Dataset info from /home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101
-INFO:datasets.builder:Found cached dataset text (/home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101)
-INFO:datasets.info:Loading Dataset info from /home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101
-INFO:datasets.arrow_dataset:Loading cached processed dataset at /home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101/cache-22cb64ac8b531f65.arrow
-INFO:datasets.arrow_dataset:Loading cached processed dataset at /home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101/cache-7840ce8488671e31.arrow
+INFO:datasets.info:Loading Dataset info from /home/iais_marenpielka/.cache/huggingface/datasets/text/default-70891baac37034df/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101
+INFO:datasets.builder:Found cached dataset text (/home/iais_marenpielka/.cache/huggingface/datasets/text/default-70891baac37034df/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101)
+INFO:datasets.info:Loading Dataset info from /home/iais_marenpielka/.cache/huggingface/datasets/text/default-70891baac37034df/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101
+INFO:datasets.arrow_dataset:Loading cached processed dataset at /home/iais_marenpielka/.cache/huggingface/datasets/text/default-70891baac37034df/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101/cache-0c0084b2e51c93f9.arrow
+INFO:datasets.arrow_dataset:Loading cached processed dataset at /home/iais_marenpielka/.cache/huggingface/datasets/text/default-70891baac37034df/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101/cache-77487910604ea5b4.arrow
 WARNING:__main__:The tokenizer picked seems to have a very large `model_max_length` (1000000000000000019884624838656). Using block_size=768 instead. You can change that default value by passing --block_size xxx.
-INFO:datasets.arrow_dataset:Loading cached processed dataset at /home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101/cache-1f7e65ff1cef4012.arrow
-INFO:datasets.arrow_dataset:Loading cached processed dataset at /home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101/cache-cb43c016a75212ff.arrow
+INFO:datasets.arrow_dataset:Loading cached processed dataset at /home/iais_marenpielka/.cache/huggingface/datasets/text/default-70891baac37034df/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101/cache-f0985db2addeab61.arrow
+INFO:datasets.arrow_dataset:Loading cached processed dataset at /home/iais_marenpielka/.cache/huggingface/datasets/text/default-70891baac37034df/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101/cache-555eefcdfe4bb35c.arrow
 WARNING:accelerate.utils.other:Detected kernel version 5.4.0, which is below the recommended minimum of 5.5.0; this can cause the process to hang. It is recommended to upgrade the kernel to the minimum version or higher.
-INFO:root:Epoch 4.0: Train Loss = 0.5673, Eval Loss = 0.6498554348945618
-INFO:absl:Using default tokenizer.
-INFO:root:Epoch 5.0: Train Loss = 0.556, Eval Loss = 0.63295978307724
-INFO:absl:Using default tokenizer.
-INFO:root:Epoch 6.0: Train Loss = 0.4786, Eval Loss = 0.6265950798988342
-INFO:absl:Using default tokenizer.
-INFO:root:Epoch 7.0: Train Loss = 0.4123, Eval Loss = 0.630312979221344
-INFO:absl:Using default tokenizer.
-INFO:root:Epoch 8.0: Train Loss = 0.3573, Eval Loss = 0.6372247338294983
-INFO:absl:Using default tokenizer.
-INFO:root:Epoch 9.0: Train Loss = 0.3108, Eval Loss = 0.646577000617981
-INFO:absl:Using default tokenizer.
-INFO:root:Epoch 10.0: Train Loss = 0.2719, Eval Loss = 0.6550981998443604
-INFO:absl:Using default tokenizer.
-INFO:__main__:*** Evaluate ***
-INFO:absl:Using default tokenizer.
-WARNING:__main__:Process rank: 0, device: cuda:0, n_gpu: 1, distributed training: False, 16-bits training: False
-INFO:__main__:Training/evaluation parameters TrainingArguments(
-_n_gpu=1,
-accelerator_config={'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False},
-adafactor=False,
-adam_beta1=0.9,
-adam_beta2=0.999,
-adam_epsilon=1e-08,
-auto_find_batch_size=False,
-batch_eval_metrics=False,
-bf16=False,
-bf16_full_eval=False,
-data_seed=None,
-dataloader_drop_last=False,
-dataloader_num_workers=0,
-dataloader_persistent_workers=False,
-dataloader_pin_memory=True,
-dataloader_prefetch_factor=None,
-ddp_backend=None,
-ddp_broadcast_buffers=None,
-ddp_bucket_cap_mb=None,
-ddp_find_unused_parameters=None,
-ddp_timeout=1800,
-debug=[],
-deepspeed=None,
-disable_tqdm=False,
-dispatch_batches=None,
-do_eval=True,
-do_predict=False,
-do_train=True,
-eval_accumulation_steps=None,
-eval_delay=0,
-eval_do_concat_batches=True,
-eval_on_start=False,
-eval_steps=None,
-eval_strategy=IntervalStrategy.EPOCH,
-eval_use_gather_object=False,
-evaluation_strategy=epoch,
-fp16=False,
-fp16_backend=auto,
-fp16_full_eval=False,
-fp16_opt_level=O1,
-fsdp=[],
-fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False},
-fsdp_min_num_params=0,
-fsdp_transformer_layer_cls_to_wrap=None,
-full_determinism=False,
-gradient_accumulation_steps=1,
-gradient_checkpointing=False,
-gradient_checkpointing_kwargs=None,
-greater_is_better=False,
-group_by_length=False,
-half_precision_backend=auto,
-hub_always_push=False,
-hub_model_id=None,
-hub_private_repo=False,
-hub_strategy=HubStrategy.EVERY_SAVE,
-hub_token=<HUB_TOKEN>,
-ignore_data_skip=False,
-include_inputs_for_metrics=False,
-include_num_input_tokens_seen=False,
-include_tokens_per_second=False,
-jit_mode_eval=False,
-label_names=None,
-label_smoothing_factor=0.0,
-learning_rate=5e-05,
-length_column_name=length,
-load_best_model_at_end=True,
-local_rank=0,
-log_level=passive,
-log_level_replica=warning,
-log_on_each_node=True,
-logging_dir=/home/iais_marenpielka/Bouthaina/res_nw_dj/runs/Sep01_08-31-45_lmgpu-node-09,
-logging_first_step=False,
-logging_nan_inf_filter=True,
-logging_steps=500,
-logging_strategy=IntervalStrategy.EPOCH,
-lr_scheduler_kwargs={},
-lr_scheduler_type=SchedulerType.LINEAR,
-max_grad_norm=1.0,
-max_steps=-1,
-metric_for_best_model=loss,
-mp_parameters=,
-neftune_noise_alpha=None,
-no_cuda=False,
-num_train_epochs=20.0,
-optim=OptimizerNames.ADAMW_TORCH,
-optim_args=None,
-optim_target_modules=None,
-output_dir=/home/iais_marenpielka/Bouthaina/res_nw_dj,
-overwrite_output_dir=False,
-past_index=-1,
-per_device_eval_batch_size=8,
-per_device_train_batch_size=8,
-prediction_loss_only=False,
-push_to_hub=True,
-push_to_hub_model_id=None,
-push_to_hub_organization=None,
-push_to_hub_token=<PUSH_TO_HUB_TOKEN>,
-ray_scope=last,
-remove_unused_columns=True,
-report_to=[],
-restore_callback_states_from_checkpoint=False,
-resume_from_checkpoint=None,
-run_name=/home/iais_marenpielka/Bouthaina/res_nw_dj,
-save_on_each_node=False,
-save_only_model=False,
-save_safetensors=True,
-save_steps=500,
-save_strategy=IntervalStrategy.EPOCH,
-save_total_limit=None,
-seed=42,
-skip_memory_metrics=True,
-split_batches=None,
-tf32=None,
-torch_compile=False,
-torch_compile_backend=None,
-torch_compile_mode=None,
-torch_empty_cache_steps=None,
-torchdynamo=None,
-tpu_metrics_debug=False,
-tpu_num_cores=None,
-use_cpu=False,
-use_ipex=False,
-use_legacy_prediction_loop=False,
-use_mps_device=False,
-warmup_ratio=0.0,
-warmup_steps=500,
-weight_decay=0.0,
-)
-INFO:__main__:Checkpoint detected, resuming training at /home/iais_marenpielka/Bouthaina/res_nw_dj/checkpoint-27030. To avoid this behavior, change the `--output_dir` or add `--overwrite_output_dir` to train from scratch.
-INFO:datasets.builder:Using custom data configuration default-98487e126fdb56c7
-INFO:datasets.info:Loading Dataset Infos from /home/iais_marenpielka/Bouthaina/miniconda3/lib/python3.12/site-packages/datasets/packaged_modules/text
-INFO:datasets.builder:Overwrite dataset info from restored data version if exists.
-INFO:datasets.info:Loading Dataset info from /home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101
-INFO:datasets.builder:Found cached dataset text (/home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101)
-INFO:datasets.info:Loading Dataset info from /home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101
-INFO:datasets.arrow_dataset:Loading cached processed dataset at /home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101/cache-22cb64ac8b531f65.arrow
-INFO:datasets.arrow_dataset:Loading cached processed dataset at /home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101/cache-7840ce8488671e31.arrow
-WARNING:__main__:The tokenizer picked seems to have a very large `model_max_length` (1000000000000000019884624838656). Using block_size=768 instead. You can change that default value by passing --block_size xxx.
-INFO:datasets.arrow_dataset:Loading cached processed dataset at /home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101/cache-1f7e65ff1cef4012.arrow
-INFO:datasets.arrow_dataset:Loading cached processed dataset at /home/iais_marenpielka/.cache/huggingface/datasets/text/default-98487e126fdb56c7/0.0.0/96636a050ef51804b84abbfd4f4ad440e01153c24b86293eb5c3b300a41f9101/cache-cb43c016a75212ff.arrow
-WARNING:accelerate.utils.other:Detected kernel version 5.4.0, which is below the recommended minimum of 5.5.0; this can cause the process to hang. It is recommended to upgrade the kernel to the minimum version or higher.
-INFO:root:Epoch 11.0: Train Loss = 0.2396, Eval Loss = 0.665830671787262
-INFO:absl:Using default tokenizer.
-INFO:root:Epoch 12.0: Train Loss = 0.2129, Eval Loss = 0.6767598390579224
-INFO:absl:Using default tokenizer.
-INFO:root:Epoch 13.0: Train Loss = 0.191, Eval Loss = 0.6870447993278503
-INFO:absl:Using default tokenizer.
-INFO:root:Epoch 14.0: Train Loss = 0.1733, Eval Loss = 0.696670651435852
-INFO:absl:Using default tokenizer.
-INFO:root:Epoch 15.0: Train Loss = 0.1593, Eval Loss = 0.7063180208206177
-INFO:absl:Using default tokenizer.
-INFO:__main__:*** Evaluate ***
+INFO:root:Epoch 1.0: Train Loss = None, Eval Loss = None
 INFO:absl:Using default tokenizer.

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bfe3c622bc3f36c295706ec7e3fd153d28f79268fcf6e11f09f6a77f69068610
-size 539221632
+oid sha256:e5633632cfa93c82e5f9c7a20da12a6109f68ad46e93939e3bc45b311171d6e4
+size 539224704

special_tokens_map.json

@@ -1,29 +1,22 @@
 {
   "additional_special_tokens": [
     {
-      "content": "<s>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    },
-    {
-      "content": "</s>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    },
-    {
-      "content": "[sep]",
+      "content": "<sep>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
       "single_word": false
     }
   ],
+  "bos_token": {
+    "content": "<|bos|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
   "eos_token": {
-    "content": "<EOS>",
+    "content": "<|endoftext|>",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,
@@ -35,5 +28,12 @@
     "normalized": false,
     "rstrip": false,
     "single_word": false
+  },
+  "unk_token": {
+    "content": "<|unk|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
   }
 }

tokenizer.json

@@ -46,7 +46,7 @@
     },
     {
       "id": 64002,
-      "content": "<s>",
+      "content": "<|bos|>",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -55,7 +55,7 @@
     },
     {
       "id": 64003,
-      "content": "</s>",
+      "content": "<|endoftext|>",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -64,7 +64,16 @@
     },
     {
       "id": 64004,
-      "content": "[sep]",
+      "content": "<|unk|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 64005,
+      "content": "<sep>",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,

tokenizer_config.json

@@ -25,7 +25,7 @@
       "special": true
     },
     "64002": {
-      "content": "<s>",
+      "content": "<|bos|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -33,7 +33,7 @@
       "special": true
     },
     "64003": {
-      "content": "</s>",
+      "content": "<|endoftext|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -41,7 +41,15 @@
       "special": true
     },
     "64004": {
-      "content": "[sep]",
+      "content": "<|unk|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "64005": {
+      "content": "<sep>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -50,13 +58,13 @@
     }
   },
   "additional_special_tokens": [
-    "<s>",
-    "</s>",
-    "[sep]"
+    "<sep>"
   ],
+  "bos_token": "<|bos|>",
   "clean_up_tokenization_spaces": true,
-  "eos_token": "<EOS>",
+  "eos_token": "<|endoftext|>",
   "model_max_length": 1000000000000000019884624838656,
   "pad_token": "<EOS>",
-  "tokenizer_class": "PreTrainedTokenizerFast"
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "unk_token": "<|unk|>"
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f941b19a4bdd74821701732fc2412174ac26cea61752479531dfffb1551ec828
+oid sha256:87284e9c550cf2a7848755d719d0d88168c066eb32f4022b467068d60eb7b769
 size 5240