Upload folder using huggingface_hub

checkpoint-8030/config.json

@@ -0,0 +1,58 @@
+{
+  "activation_dropout": 0.1,
+  "activation_function": "gelu",
+  "add_bias_logits": false,
+  "add_final_layer_norm": true,
+  "architectures": [
+    "MBartForConditionalGeneration"
+  ],
+  "attention_dropout": 0.1,
+  "bos_token_id": 0,
+  "classif_dropout": 0.1,
+  "classifier_dropout": 0.0,
+  "d_model": 768,
+  "decoder_attention_heads": 12,
+  "decoder_ffn_dim": 3072,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 6,
+  "decoder_start_token_id": 2,
+  "do_blenderbot_90_layernorm": false,
+  "dropout": 0.1,
+  "early_stopping": null,
+  "encoder_attention_heads": 12,
+  "encoder_ffn_dim": 3072,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 6,
+  "eos_token_id": 2,
+  "extra_pos_embeddings": 2,
+  "force_bos_token_to_be_generated": false,
+  "forced_eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "max_position_embeddings": 1024,
+  "model_type": "mbart",
+  "no_repeat_ngram_size": null,
+  "normalize_before": true,
+  "normalize_embedding": true,
+  "num_beams": null,
+  "num_hidden_layers": 6,
+  "pad_token_id": 1,
+  "scale_embedding": false,
+  "static_position_embeddings": false,
+  "tokenizer_class": "BarthezTokenizer",
+  "torch_dtype": "float32",
+  "transformers_version": "4.51.3",
+  "use_cache": true,
+  "vocab_size": 86001
+}

checkpoint-8030/generation_config.json

@@ -0,0 +1,8 @@
+{
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "eos_token_id": 2,
+  "forced_eos_token_id": 2,
+  "pad_token_id": 1,
+  "transformers_version": "4.51.3"
+}

checkpoint-8030/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7d9179be7efa3d039286efbd45485bca053994cd099d18d9332ecc07e690565
+size 667849236

checkpoint-8030/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:670d6167cb7bd797530de458573b94b7d6cfcd54d0e732454c3ec913c2f5df18
+size 1964811

checkpoint-8030/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b5c9a321d911642f4f39e094bc71f7f606a7c0729b70323a4e97b3ed73f476b
+size 14645

checkpoint-8030/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:788952817d7e20534058a75dacbdd9323070be90cfa509634b49d011404572c0
+size 1383

checkpoint-8030/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a317addef3c200d7a4f262c74fdc4a21ceba55a5883e574984d5309cb5ed2ad9
+size 1465

checkpoint-8030/special_tokens_map.json

@@ -0,0 +1,9 @@
+{
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-8030/tokenizer_config.json

@@ -0,0 +1,25 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "86000": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "extra_special_tokens": {},
+  "model_max_length": 128,
+  "pad_token": "[PAD]",
+  "tokenizer_class": "PreTrainedTokenizer"
+}

checkpoint-8030/trainer_state.json

@@ -0,0 +1,206 @@
+{
+  "best_global_step": 8030,
+  "best_metric": 0.7079419812877328,
+  "best_model_checkpoint": "checkpoints/checkpoint-8030",
+  "epoch": 1.2499902712378876,
+  "eval_steps": 1606,
+  "global_step": 8030,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.07783009689847063,
+      "grad_norm": 0.2925865948200226,
+      "learning_rate": 4.991766654390158e-05,
+      "loss": 2.2108,
+      "step": 500
+    },
+    {
+      "epoch": 0.15566019379694127,
+      "grad_norm": 0.11884481459856033,
+      "learning_rate": 4.966988962239142e-05,
+      "loss": 0.3886,
+      "step": 1000
+    },
+    {
+      "epoch": 0.2334902906954119,
+      "grad_norm": 0.09392867982387543,
+      "learning_rate": 4.925831306092169e-05,
+      "loss": 0.2271,
+      "step": 1500
+    },
+    {
+      "epoch": 0.2499902712378877,
+      "eval_avg": 26.749637281629344,
+      "eval_cer": 0.6640905490705347,
+      "eval_der": 78.87487905513909,
+      "eval_loss": 0.13775022327899933,
+      "eval_runtime": 1439.866,
+      "eval_samples_per_second": 94.261,
+      "eval_steps_per_second": 0.369,
+      "eval_wer": 0.7099422406784007,
+      "step": 1606
+    },
+    {
+      "epoch": 0.31132038759388253,
+      "grad_norm": 0.0717502012848854,
+      "learning_rate": 4.8685669587267704e-05,
+      "loss": 0.1657,
+      "step": 2000
+    },
+    {
+      "epoch": 0.3891504844923532,
+      "grad_norm": 0.0638226792216301,
+      "learning_rate": 4.795576135848184e-05,
+      "loss": 0.1334,
+      "step": 2500
+    },
+    {
+      "epoch": 0.4669805813908238,
+      "grad_norm": 0.05441045016050339,
+      "learning_rate": 4.707343471586959e-05,
+      "loss": 0.1135,
+      "step": 3000
+    },
+    {
+      "epoch": 0.4999805424757754,
+      "eval_avg": 26.867260983703588,
+      "eval_cer": 0.6597599235984197,
+      "eval_der": 79.24906581788363,
+      "eval_loss": 0.07722582668066025,
+      "eval_runtime": 1442.4649,
+      "eval_samples_per_second": 94.091,
+      "eval_steps_per_second": 0.368,
+      "eval_wer": 0.6929572096287142,
+      "step": 3212
+    },
+    {
+      "epoch": 0.5448106782892945,
+      "grad_norm": 0.050636373460292816,
+      "learning_rate": 4.604454800693874e-05,
+      "loss": 0.1002,
+      "step": 3500
+    },
+    {
+      "epoch": 0.6226407751877651,
+      "grad_norm": 0.0581156350672245,
+      "learning_rate": 4.48759326879731e-05,
+      "loss": 0.0906,
+      "step": 4000
+    },
+    {
+      "epoch": 0.7004708720862357,
+      "grad_norm": 0.050500743091106415,
+      "learning_rate": 4.3575347965496405e-05,
+      "loss": 0.0828,
+      "step": 4500
+    },
+    {
+      "epoch": 0.749970813713663,
+      "eval_avg": 26.888053782081823,
+      "eval_cer": 0.6586451367564623,
+      "eval_der": 79.3192047037025,
+      "eval_loss": 0.05902250111103058,
+      "eval_runtime": 1431.2383,
+      "eval_samples_per_second": 94.829,
+      "eval_steps_per_second": 0.371,
+      "eval_wer": 0.6863115057865005,
+      "step": 4818
+    },
+    {
+      "epoch": 0.7783009689847064,
+      "grad_norm": 0.04594194516539574,
+      "learning_rate": 4.9989118673869795e-05,
+      "loss": 0.0765,
+      "step": 5000
+    },
+    {
+      "epoch": 0.856131065883177,
+      "grad_norm": 0.04987065866589546,
+      "learning_rate": 4.984611214237347e-05,
+      "loss": 0.0724,
+      "step": 5500
+    },
+    {
+      "epoch": 0.9339611627816476,
+      "grad_norm": 0.04362853616476059,
+      "learning_rate": 4.953813591358179e-05,
+      "loss": 0.0683,
+      "step": 6000
+    },
+    {
+      "epoch": 0.9999610849515508,
+      "eval_avg": 0.708712044715695,
+      "eval_cer": 0.6546741642910598,
+      "eval_der": 0.7934032551594553,
+      "eval_loss": 0.049141544848680496,
+      "eval_runtime": 305.8639,
+      "eval_samples_per_second": 443.737,
+      "eval_steps_per_second": 1.736,
+      "eval_wer": 0.6780587146965699,
+      "step": 6424
+    },
+    {
+      "epoch": 1.0118301747285676,
+      "grad_norm": 0.04417261481285095,
+      "learning_rate": 4.99981315851881e-05,
+      "loss": 0.0626,
+      "step": 6500
+    },
+    {
+      "epoch": 1.0896602716270383,
+      "grad_norm": 0.039167486131191254,
+      "learning_rate": 4.9890257744619245e-05,
+      "loss": 0.061,
+      "step": 7000
+    },
+    {
+      "epoch": 1.1674903685255087,
+      "grad_norm": 0.035239290446043015,
+      "learning_rate": 4.961712109503609e-05,
+      "loss": 0.0583,
+      "step": 7500
+    },
+    {
+      "epoch": 1.2453204654239793,
+      "grad_norm": 0.03707383945584297,
+      "learning_rate": 4.918053517048949e-05,
+      "loss": 0.056,
+      "step": 8000
+    },
+    {
+      "epoch": 1.2499902712378876,
+      "eval_avg": 0.7079419812877328,
+      "eval_cer": 0.6543333833811898,
+      "eval_der": 0.793713077370868,
+      "eval_loss": 0.043703265488147736,
+      "eval_runtime": 307.3798,
+      "eval_samples_per_second": 441.548,
+      "eval_steps_per_second": 1.728,
+      "eval_wer": 0.6757794831111406,
+      "step": 8030
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 19272,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 1606,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 6.267309210678067e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-8030/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13cd8f43a255510665b26c3fb2a6d437f327d011dfaac3168ef02538a3bb5da9
+size 5905

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:74bcde2a9eb993ebfa7ad2120a125469366ea72d5b50e4642e9ab73eaec297c6
+oid sha256:d7d9179be7efa3d039286efbd45485bca053994cd099d18d9332ecc07e690565
 size 667849236

runs/May05_22-12-51_ip-10-192-11-107/events.out.tfevents.1746483172.ip-10-192-11-107.1742.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:136f8fa20bb71a3989194a73774cece3c3981c420128d947f5cc4475711d16a7
+size 5938

runs/May05_22-15-18_ip-10-192-11-107/events.out.tfevents.1746483320.ip-10-192-11-107.1742.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4fa1b1793cd7a32a78d0adb81f839c2d36b48f7ef40f2b34aab870ba9924d01c
+size 7241

trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 6424,
-  "best_metric": 0.708712044715695,
-  "best_model_checkpoint": "checkpoints/checkpoint-6424",
-  "epoch": 0.9999610849515508,
+  "best_global_step": 8030,
+  "best_metric": 0.7079419812877328,
+  "best_model_checkpoint": "checkpoints/checkpoint-8030",
+  "epoch": 1.2499902712378876,
   "eval_steps": 1606,
-  "global_step": 6424,
+  "global_step": 8030,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -140,6 +140,46 @@
       "eval_steps_per_second": 1.736,
       "eval_wer": 0.6780587146965699,
       "step": 6424
+    },
+    {
+      "epoch": 1.0118301747285676,
+      "grad_norm": 0.04417261481285095,
+      "learning_rate": 4.99981315851881e-05,
+      "loss": 0.0626,
+      "step": 6500
+    },
+    {
+      "epoch": 1.0896602716270383,
+      "grad_norm": 0.039167486131191254,
+      "learning_rate": 4.9890257744619245e-05,
+      "loss": 0.061,
+      "step": 7000
+    },
+    {
+      "epoch": 1.1674903685255087,
+      "grad_norm": 0.035239290446043015,
+      "learning_rate": 4.961712109503609e-05,
+      "loss": 0.0583,
+      "step": 7500
+    },
+    {
+      "epoch": 1.2453204654239793,
+      "grad_norm": 0.03707383945584297,
+      "learning_rate": 4.918053517048949e-05,
+      "loss": 0.056,
+      "step": 8000
+    },
+    {
+      "epoch": 1.2499902712378876,
+      "eval_avg": 0.7079419812877328,
+      "eval_cer": 0.6543333833811898,
+      "eval_der": 0.793713077370868,
+      "eval_loss": 0.043703265488147736,
+      "eval_runtime": 307.3798,
+      "eval_samples_per_second": 441.548,
+      "eval_steps_per_second": 1.728,
+      "eval_wer": 0.6757794831111406,
+      "step": 8030
     }
   ],
   "logging_steps": 500,
@@ -159,7 +199,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 5.013847368542454e+17,
+  "total_flos": 6.267309210678067e+17,
   "train_batch_size": 256,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1cd77b85635f40803f2a499bdb589f69718a7b080ce297cef7adac845287846e
+oid sha256:13cd8f43a255510665b26c3fb2a6d437f327d011dfaac3168ef02538a3bb5da9
 size 5905