Next training iteration (290k)

added_tokens.json

@@ -1 +1 @@
-{"START_INTERACTION_TOKEN": 32116, "LONG_WORD_TOKEN": 32106, "NUMBER_PERCENTAGE_TOKEN": 32103, "END_INTERACTION_TOKEN": 32117, "START_SELFPROMO_TOKEN": 32114, "[Music]": 32107, "END_SPONSOR_TOKEN": 32113, "URL_TOKEN": 32101, "NUMBER_TOKEN": 32104, "PROFANITY_TOKEN": 32110, "HYPHENATED_URL_TOKEN": 32102, "END_SELFPROMO_TOKEN": 32115, "NO_SEGMENT_TOKEN": 32111, "SHORT_HYPHENATED_TOKEN": 32105, "BETWEEN_SEGMENTS_TOKEN": 32118, "START_SPONSOR_TOKEN": 32112, "EXTRACT_SEGMENTS: ": 32100, "[Applause]": 32108, "[Laughter]": 32109}
+{"[Applause]": 32108, "START_INTERACTION_TOKEN": 32116, "NUMBER_TOKEN": 32104, "PROFANITY_TOKEN": 32110, "START_SELFPROMO_TOKEN": 32114, "EXTRACT_SEGMENTS: ": 32100, "END_SPONSOR_TOKEN": 32113, "END_SELFPROMO_TOKEN": 32115, "URL_TOKEN": 32101, "NUMBER_PERCENTAGE_TOKEN": 32103, "END_INTERACTION_TOKEN": 32117, "BETWEEN_SEGMENTS_TOKEN": 32118, "LONG_WORD_TOKEN": 32106, "[Laughter]": 32109, "HYPHENATED_URL_TOKEN": 32102, "START_SPONSOR_TOKEN": 32112, "[Music]": 32107, "NO_SEGMENT_TOKEN": 32111, "SHORT_HYPHENATED_TOKEN": 32105}

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "google/t5-v1_1-base",
+  "_name_or_path": "Xenova/sponsorblock-base-v1.1",
   "architectures": [
     "T5ForConditionalGeneration"
   ],
@@ -22,7 +22,7 @@
   "relative_attention_num_buckets": 32,
   "tie_word_embeddings": false,
   "torch_dtype": "float32",
-  "transformers_version": "4.16.1",
+  "transformers_version": "4.16.2",
   "use_cache": true,
   "vocab_size": 32119
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a23fa3d63f7b869dd2fcde0ae59b6f84a4d6a7c2e5e7cea918425d09eec82f7a
+oid sha256:1de277f5eccd1a9682d02507d2f73c334c91299bf453697b554e9a853695c8c8
 size 990383053

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c730413130fc63756f54d5820e7a7171dce6b48472fe73812d2eeee7d3ecc44d
+oid sha256:0a7ce493844e3989ca53d1f40d8eacf32538194607557f372b49129909070e97
 size 14503

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8fc17a095b50cb3c162e9244c249edb34b9632ba405100031ff659035a9789a7
+oid sha256:273a7d9a6cd9cbcd2571b4ecb3d1f0cbe8dbb6916d43a93fc7a7afd46a46c400
 size 623

tokenizer_config.json

@@ -1 +1 @@
-{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "extra_ids": 100, "additional_special_tokens": ["<extra_id_0>", "<extra_id_1>", "<extra_id_2>", "<extra_id_3>", "<extra_id_4>", "<extra_id_5>", "<extra_id_6>", "<extra_id_7>", "<extra_id_8>", "<extra_id_9>", "<extra_id_10>", "<extra_id_11>", "<extra_id_12>", "<extra_id_13>", "<extra_id_14>", "<extra_id_15>", "<extra_id_16>", "<extra_id_17>", "<extra_id_18>", "<extra_id_19>", "<extra_id_20>", "<extra_id_21>", "<extra_id_22>", "<extra_id_23>", "<extra_id_24>", "<extra_id_25>", "<extra_id_26>", "<extra_id_27>", "<extra_id_28>", "<extra_id_29>", "<extra_id_30>", "<extra_id_31>", "<extra_id_32>", "<extra_id_33>", "<extra_id_34>", "<extra_id_35>", "<extra_id_36>", "<extra_id_37>", "<extra_id_38>", "<extra_id_39>", "<extra_id_40>", "<extra_id_41>", "<extra_id_42>", "<extra_id_43>", "<extra_id_44>", "<extra_id_45>", "<extra_id_46>", "<extra_id_47>", "<extra_id_48>", "<extra_id_49>", "<extra_id_50>", "<extra_id_51>", "<extra_id_52>", "<extra_id_53>", "<extra_id_54>", "<extra_id_55>", "<extra_id_56>", "<extra_id_57>", "<extra_id_58>", "<extra_id_59>", "<extra_id_60>", "<extra_id_61>", "<extra_id_62>", "<extra_id_63>", "<extra_id_64>", "<extra_id_65>", "<extra_id_66>", "<extra_id_67>", "<extra_id_68>", "<extra_id_69>", "<extra_id_70>", "<extra_id_71>", "<extra_id_72>", "<extra_id_73>", "<extra_id_74>", "<extra_id_75>", "<extra_id_76>", "<extra_id_77>", "<extra_id_78>", "<extra_id_79>", "<extra_id_80>", "<extra_id_81>", "<extra_id_82>", "<extra_id_83>", "<extra_id_84>", "<extra_id_85>", "<extra_id_86>", "<extra_id_87>", "<extra_id_88>", "<extra_id_89>", "<extra_id_90>", "<extra_id_91>", "<extra_id_92>", "<extra_id_93>", "<extra_id_94>", "<extra_id_95>", "<extra_id_96>", "<extra_id_97>", "<extra_id_98>", "<extra_id_99>"], "model_max_length": 512, "name_or_path": "google/t5-v1_1-base", "max_length": 768, "special_tokens_map_file": "/root/.cache/huggingface/transformers/76bf19bfedb85afbe644966ca9ab7b0404d753a41bf601115bced39f825ffa9c.c94798918c92ded6aeef2d2f0e666d2cc4145eca1aa6e1336fde07f2e13e2f46", "sp_model_kwargs": {}, "tokenizer_class": "T5Tokenizer"}
+{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "extra_ids": 100, "additional_special_tokens": ["<extra_id_0>", "<extra_id_1>", "<extra_id_2>", "<extra_id_3>", "<extra_id_4>", "<extra_id_5>", "<extra_id_6>", "<extra_id_7>", "<extra_id_8>", "<extra_id_9>", "<extra_id_10>", "<extra_id_11>", "<extra_id_12>", "<extra_id_13>", "<extra_id_14>", "<extra_id_15>", "<extra_id_16>", "<extra_id_17>", "<extra_id_18>", "<extra_id_19>", "<extra_id_20>", "<extra_id_21>", "<extra_id_22>", "<extra_id_23>", "<extra_id_24>", "<extra_id_25>", "<extra_id_26>", "<extra_id_27>", "<extra_id_28>", "<extra_id_29>", "<extra_id_30>", "<extra_id_31>", "<extra_id_32>", "<extra_id_33>", "<extra_id_34>", "<extra_id_35>", "<extra_id_36>", "<extra_id_37>", "<extra_id_38>", "<extra_id_39>", "<extra_id_40>", "<extra_id_41>", "<extra_id_42>", "<extra_id_43>", "<extra_id_44>", "<extra_id_45>", "<extra_id_46>", "<extra_id_47>", "<extra_id_48>", "<extra_id_49>", "<extra_id_50>", "<extra_id_51>", "<extra_id_52>", "<extra_id_53>", "<extra_id_54>", "<extra_id_55>", "<extra_id_56>", "<extra_id_57>", "<extra_id_58>", "<extra_id_59>", "<extra_id_60>", "<extra_id_61>", "<extra_id_62>", "<extra_id_63>", "<extra_id_64>", "<extra_id_65>", "<extra_id_66>", "<extra_id_67>", "<extra_id_68>", "<extra_id_69>", "<extra_id_70>", "<extra_id_71>", "<extra_id_72>", "<extra_id_73>", "<extra_id_74>", "<extra_id_75>", "<extra_id_76>", "<extra_id_77>", "<extra_id_78>", "<extra_id_79>", "<extra_id_80>", "<extra_id_81>", "<extra_id_82>", "<extra_id_83>", "<extra_id_84>", "<extra_id_85>", "<extra_id_86>", "<extra_id_87>", "<extra_id_88>", "<extra_id_89>", "<extra_id_90>", "<extra_id_91>", "<extra_id_92>", "<extra_id_93>", "<extra_id_94>", "<extra_id_95>", "<extra_id_96>", "<extra_id_97>", "<extra_id_98>", "<extra_id_99>"], "model_max_length": 512, "name_or_path": "Xenova/sponsorblock-base-v1.1", "max_length": 768, "special_tokens_map_file": "/root/.cache/huggingface/transformers/76bf19bfedb85afbe644966ca9ab7b0404d753a41bf601115bced39f825ffa9c.c94798918c92ded6aeef2d2f0e666d2cc4145eca1aa6e1336fde07f2e13e2f46", "sp_model_kwargs": {}, "tokenizer_class": "T5Tokenizer"}

trainer_state.json

@@ -1,263 +1,174 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.0634139611882645,
-  "global_step": 185000,
+  "epoch": 0.31412202892914304,
+  "global_step": 105000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.01,
-      "learning_rate": 4.955220966618126e-05,
-      "loss": 2.3776,
+      "learning_rate": 4.925209040731157e-05,
+      "loss": 0.0716,
       "step": 5000
     },
     {
-      "epoch": 0.02,
-      "learning_rate": 4.910441933236253e-05,
-      "loss": 0.2443,
+      "epoch": 0.03,
+      "learning_rate": 4.850418081462313e-05,
+      "loss": 0.0741,
       "step": 10000
     },
     {
-      "epoch": 0.03,
-      "learning_rate": 4.865662899854379e-05,
-      "loss": 0.2738,
+      "epoch": 0.04,
+      "learning_rate": 4.77562712219347e-05,
+      "loss": 0.0753,
       "step": 15000
     },
     {
-      "epoch": 0.04,
-      "learning_rate": 4.8208838664725045e-05,
-      "loss": 0.1848,
+      "epoch": 0.06,
+      "learning_rate": 4.7008361629246264e-05,
+      "loss": 0.0709,
       "step": 20000
     },
     {
-      "epoch": 0.04,
-      "learning_rate": 4.776104833090631e-05,
-      "loss": 0.1507,
+      "epoch": 0.07,
+      "learning_rate": 4.6260452036557824e-05,
+      "loss": 0.0754,
       "step": 25000
     },
     {
-      "epoch": 0.04,
-      "eval_loss": 0.11242648959159851,
-      "eval_runtime": 3430.1226,
-      "eval_samples_per_second": 18.085,
-      "eval_steps_per_second": 9.043,
+      "epoch": 0.07,
+      "eval_loss": 0.07123171538114548,
+      "eval_runtime": 1351.5318,
+      "eval_samples_per_second": 13.741,
+      "eval_steps_per_second": 13.741,
       "step": 25000
     },
     {
-      "epoch": 0.05,
-      "learning_rate": 4.7313257997087574e-05,
-      "loss": 0.1358,
+      "epoch": 0.09,
+      "learning_rate": 4.551254244386939e-05,
+      "loss": 0.0745,
       "step": 30000
     },
     {
-      "epoch": 0.06,
-      "learning_rate": 4.6865467663268834e-05,
-      "loss": 0.1323,
+      "epoch": 0.1,
+      "learning_rate": 4.476463285118095e-05,
+      "loss": 0.0671,
       "step": 35000
     },
     {
-      "epoch": 0.07,
-      "learning_rate": 4.64176773294501e-05,
-      "loss": 0.1193,
+      "epoch": 0.12,
+      "learning_rate": 4.401672325849251e-05,
+      "loss": 0.0687,
       "step": 40000
     },
     {
-      "epoch": 0.08,
-      "learning_rate": 4.596988699563136e-05,
-      "loss": 0.1129,
+      "epoch": 0.13,
+      "learning_rate": 4.326881366580408e-05,
+      "loss": 0.0718,
       "step": 45000
     },
     {
-      "epoch": 0.09,
-      "learning_rate": 4.5522096661812623e-05,
-      "loss": 0.1115,
+      "epoch": 0.15,
+      "learning_rate": 4.252090407311564e-05,
+      "loss": 0.0692,
       "step": 50000
     },
-    {
-      "epoch": 0.1,
-      "learning_rate": 4.5074306327993884e-05,
-      "loss": 0.1072,
-      "step": 55000
-    },
-    {
-      "epoch": 0.11,
-      "learning_rate": 4.4626515994175145e-05,
-      "loss": 0.1079,
-      "step": 60000
-    },
-    {
-      "epoch": 0.12,
-      "learning_rate": 4.4178725660356406e-05,
-      "loss": 0.3569,
-      "step": 65000
-    },
-    {
-      "epoch": 0.13,
-      "learning_rate": 4.3730935326537667e-05,
-      "loss": 0.1283,
-      "step": 70000
-    },
-    {
-      "epoch": 0.13,
-      "learning_rate": 4.3283144992718934e-05,
-      "loss": 0.1039,
-      "step": 75000
-    },
-    {
-      "epoch": 0.14,
-      "learning_rate": 4.2835354658900195e-05,
-      "loss": 0.0956,
-      "step": 80000
-    },
     {
       "epoch": 0.15,
-      "learning_rate": 4.2387564325081456e-05,
-      "loss": 0.0989,
-      "step": 85000
+      "eval_loss": 0.06770400702953339,
+      "eval_runtime": 1359.9214,
+      "eval_samples_per_second": 13.656,
+      "eval_steps_per_second": 13.656,
+      "step": 50000
     },
     {
       "epoch": 0.16,
-      "learning_rate": 4.1939773991262716e-05,
-      "loss": 0.0966,
-      "step": 90000
-    },
-    {
-      "epoch": 0.17,
-      "learning_rate": 4.149198365744398e-05,
-      "loss": 0.096,
-      "step": 95000
-    },
-    {
-      "epoch": 0.18,
-      "learning_rate": 4.104419332362524e-05,
-      "loss": 0.0891,
-      "step": 100000
+      "learning_rate": 4.1772994480427206e-05,
+      "loss": 0.0695,
+      "step": 55000
     },
     {
       "epoch": 0.18,
-      "eval_loss": 0.08249569684267044,
-      "eval_runtime": 3211.3155,
-      "eval_samples_per_second": 19.317,
-      "eval_steps_per_second": 9.659,
-      "step": 100000
+      "learning_rate": 4.102508488773877e-05,
+      "loss": 0.0696,
+      "step": 60000
     },
     {
       "epoch": 0.19,
-      "learning_rate": 4.0596402989806505e-05,
-      "loss": 0.0897,
-      "step": 105000
-    },
-    {
-      "epoch": 0.2,
-      "learning_rate": 4.0148612655987766e-05,
-      "loss": 0.0895,
-      "step": 110000
-    },
-    {
-      "epoch": 0.21,
-      "learning_rate": 3.970082232216903e-05,
-      "loss": 0.0929,
-      "step": 115000
+      "learning_rate": 4.0277175295050333e-05,
+      "loss": 0.0689,
+      "step": 65000
     },
     {
       "epoch": 0.21,
-      "learning_rate": 3.925303198835029e-05,
-      "loss": 0.0857,
-      "step": 120000
+      "learning_rate": 3.95292657023619e-05,
+      "loss": 0.0679,
+      "step": 70000
     },
     {
       "epoch": 0.22,
-      "learning_rate": 3.880524165453155e-05,
-      "loss": 0.0879,
-      "step": 125000
+      "learning_rate": 3.878135610967347e-05,
+      "loss": 0.071,
+      "step": 75000
     },
     {
-      "epoch": 0.23,
-      "learning_rate": 3.835745132071281e-05,
-      "loss": 0.0855,
-      "step": 130000
+      "epoch": 0.22,
+      "eval_loss": 0.05809599161148071,
+      "eval_runtime": 2705.4449,
+      "eval_samples_per_second": 6.864,
+      "eval_steps_per_second": 6.864,
+      "step": 75000
     },
     {
       "epoch": 0.24,
-      "learning_rate": 3.790966098689408e-05,
-      "loss": 0.0841,
-      "step": 135000
+      "learning_rate": 3.803344651698503e-05,
+      "loss": 0.0712,
+      "step": 80000
     },
     {
       "epoch": 0.25,
-      "learning_rate": 3.746187065307534e-05,
-      "loss": 0.0866,
-      "step": 140000
-    },
-    {
-      "epoch": 0.26,
-      "learning_rate": 3.70140803192566e-05,
-      "loss": 0.0855,
-      "step": 145000
+      "learning_rate": 3.7285536924296595e-05,
+      "loss": 0.0683,
+      "step": 85000
     },
     {
       "epoch": 0.27,
-      "learning_rate": 3.6566289985437866e-05,
-      "loss": 0.0825,
-      "step": 150000
+      "learning_rate": 3.653762733160816e-05,
+      "loss": 0.0673,
+      "step": 90000
     },
     {
       "epoch": 0.28,
-      "learning_rate": 3.611849965161912e-05,
-      "loss": 0.082,
-      "step": 155000
-    },
-    {
-      "epoch": 0.29,
-      "learning_rate": 3.567070931780038e-05,
-      "loss": 0.0865,
-      "step": 160000
+      "learning_rate": 3.578971773891972e-05,
+      "loss": 0.0685,
+      "step": 95000
     },
     {
       "epoch": 0.3,
-      "learning_rate": 3.522291898398165e-05,
-      "loss": 0.0814,
-      "step": 165000
-    },
-    {
-      "epoch": 0.98,
-      "learning_rate": 1.1404396210797388e-06,
-      "loss": 0.0811,
-      "step": 170000
-    },
-    {
-      "epoch": 1.0,
-      "step": 173968,
-      "total_flos": 1.727613380688722e+17,
-      "train_loss": 0.004168222484029031,
-      "train_runtime": 9992.3727,
-      "train_samples_per_second": 34.82,
-      "train_steps_per_second": 17.41
-    },
-    {
-      "epoch": 1.01,
-      "learning_rate": 4.997033937275821e-05,
-      "loss": 0.079,
-      "step": 175000
+      "learning_rate": 3.504180814623129e-05,
+      "loss": 0.0676,
+      "step": 100000
     },
     {
-      "epoch": 1.03,
-      "learning_rate": 4.982663478340844e-05,
-      "loss": 0.0806,
-      "step": 180000
+      "epoch": 0.3,
+      "eval_loss": 0.06122186779975891,
+      "eval_runtime": 2772.9648,
+      "eval_samples_per_second": 6.697,
+      "eval_steps_per_second": 6.697,
+      "step": 100000
     },
     {
-      "epoch": 1.06,
-      "learning_rate": 4.968293019405868e-05,
-      "loss": 0.0784,
-      "step": 185000
+      "epoch": 0.31,
+      "learning_rate": 3.429389855354285e-05,
+      "loss": 0.0672,
+      "step": 105000
     }
   ],
-  "max_steps": 1739680,
-  "num_train_epochs": 10,
-  "total_flos": 1.849684070324828e+17,
+  "max_steps": 334265,
+  "num_train_epochs": 1,
+  "total_flos": 6.792868897849498e+16,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cbfa8ba174e9fc1e47a9566b4253523ff02ac3c21b714c241b7469d42c08648c
+oid sha256:1a0c1e03198149a91f09ff9053eb5c84a9d581ebee4ffac70490d350b10a0f7d
 size 3183