Training in progress, epoch 17, checkpoint

checkpoint-3604/README.md

@@ -0,0 +1,9 @@
+---
+library_name: peft
+---
+## Training procedure
+
+### Framework versions
+
+
+- PEFT 0.5.0

checkpoint-3604/adapter_config.json

@@ -0,0 +1,26 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "camembert/camembert-large",
+  "bias": "all",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 16,
+  "lora_dropout": 0.1,
+  "modules_to_save": [
+    "classifier.bias",
+    "classifier.weight"
+  ],
+  "peft_type": "LORA",
+  "r": 64,
+  "revision": null,
+  "target_modules": [
+    "query",
+    "key",
+    "value",
+    "dense"
+  ],
+  "task_type": "TOKEN_CLS"
+}

checkpoint-3604/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:31affe527e7fa9bd392e7185bcc626e5ac7866ac592000b98387508a9092bc0d
+size 117304221

checkpoint-3604/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02b6354f9e63d6745ca61366d78fd3f7981949a494451c7135f0c03e5c7c1d36
+size 234652550

checkpoint-3604/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:030fc7a02c1ed353e753fcb840416f04602aa301edbbc4137599fd673b30722c
+size 14575

checkpoint-3604/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72df5c7319f2c653866c80fb51975e9665f5500052240eb99a0a5e66e9557995
+size 627

checkpoint-3604/sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f98f266fdc548c94216aaadc13ffaaafacf0c8793303e2195322d954549ea261
+size 808767

checkpoint-3604/special_tokens_map.json

@@ -0,0 +1,19 @@
+{
+  "additional_special_tokens": [
+    "<s>NOTUSED",
+    "</s>NOTUSED"
+  ],
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

checkpoint-3604/tokenizer_config.json

@@ -0,0 +1,24 @@
+{
+  "additional_special_tokens": [
+    "<s>NOTUSED",
+    "</s>NOTUSED"
+  ],
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "__type": "AddedToken",
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "CamembertTokenizer",
+  "unk_token": "<unk>"
+}

checkpoint-3604/trainer_state.json

@@ -0,0 +1,257 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 17.0,
+  "eval_steps": 500,
+  "global_step": 3604,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00019900471698113207,
+      "loss": 6.1329,
+      "step": 212
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 5.14426851272583,
+      "eval_runtime": 3.8194,
+      "eval_samples_per_second": 176.991,
+      "eval_steps_per_second": 22.255,
+      "step": 212
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.00019800471698113207,
+      "loss": 4.6548,
+      "step": 424
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 3.559998035430908,
+      "eval_runtime": 3.8713,
+      "eval_samples_per_second": 174.616,
+      "eval_steps_per_second": 21.956,
+      "step": 424
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.00019700471698113207,
+      "loss": 3.2979,
+      "step": 636
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.3220388889312744,
+      "eval_runtime": 3.7585,
+      "eval_samples_per_second": 179.86,
+      "eval_steps_per_second": 22.616,
+      "step": 636
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 0.00019600471698113207,
+      "loss": 2.2394,
+      "step": 848
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.5198729038238525,
+      "eval_runtime": 3.8733,
+      "eval_samples_per_second": 174.529,
+      "eval_steps_per_second": 21.945,
+      "step": 848
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.00019500471698113208,
+      "loss": 1.5436,
+      "step": 1060
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 1.0844509601593018,
+      "eval_runtime": 3.7882,
+      "eval_samples_per_second": 178.45,
+      "eval_steps_per_second": 22.438,
+      "step": 1060
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 0.0001940047169811321,
+      "loss": 1.1087,
+      "step": 1272
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 0.8275050520896912,
+      "eval_runtime": 3.8596,
+      "eval_samples_per_second": 175.146,
+      "eval_steps_per_second": 22.023,
+      "step": 1272
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 0.00019300943396226416,
+      "loss": 0.8094,
+      "step": 1484
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 0.6410268545150757,
+      "eval_runtime": 3.826,
+      "eval_samples_per_second": 176.686,
+      "eval_steps_per_second": 22.216,
+      "step": 1484
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 0.00019200943396226417,
+      "loss": 0.5999,
+      "step": 1696
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 0.5242534279823303,
+      "eval_runtime": 3.7888,
+      "eval_samples_per_second": 178.42,
+      "eval_steps_per_second": 22.434,
+      "step": 1696
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 0.00019100943396226417,
+      "loss": 0.4333,
+      "step": 1908
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 0.4525991678237915,
+      "eval_runtime": 3.7762,
+      "eval_samples_per_second": 179.014,
+      "eval_steps_per_second": 22.509,
+      "step": 1908
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 0.00019000943396226417,
+      "loss": 0.3096,
+      "step": 2120
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 0.3988202214241028,
+      "eval_runtime": 3.3771,
+      "eval_samples_per_second": 200.169,
+      "eval_steps_per_second": 25.169,
+      "step": 2120
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 0.00018900943396226415,
+      "loss": 0.225,
+      "step": 2332
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 0.37283995747566223,
+      "eval_runtime": 3.3836,
+      "eval_samples_per_second": 199.788,
+      "eval_steps_per_second": 25.121,
+      "step": 2332
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 0.00018800943396226415,
+      "loss": 0.1484,
+      "step": 2544
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 0.36433306336402893,
+      "eval_runtime": 3.9173,
+      "eval_samples_per_second": 172.567,
+      "eval_steps_per_second": 21.698,
+      "step": 2544
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 0.00018700943396226415,
+      "loss": 0.1086,
+      "step": 2756
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 0.3588293790817261,
+      "eval_runtime": 3.3573,
+      "eval_samples_per_second": 201.351,
+      "eval_steps_per_second": 25.318,
+      "step": 2756
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 0.00018600943396226415,
+      "loss": 0.0794,
+      "step": 2968
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 0.34358447790145874,
+      "eval_runtime": 3.5044,
+      "eval_samples_per_second": 192.9,
+      "eval_steps_per_second": 24.255,
+      "step": 2968
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 0.00018500943396226416,
+      "loss": 0.0629,
+      "step": 3180
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 0.3459497392177582,
+      "eval_runtime": 3.9331,
+      "eval_samples_per_second": 171.873,
+      "eval_steps_per_second": 21.611,
+      "step": 3180
+    },
+    {
+      "epoch": 16.0,
+      "learning_rate": 0.00018400943396226416,
+      "loss": 0.0499,
+      "step": 3392
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 0.35890209674835205,
+      "eval_runtime": 3.9531,
+      "eval_samples_per_second": 171.003,
+      "eval_steps_per_second": 21.502,
+      "step": 3392
+    },
+    {
+      "epoch": 17.0,
+      "learning_rate": 0.00018300943396226416,
+      "loss": 0.0381,
+      "step": 3604
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 0.36232635378837585,
+      "eval_runtime": 3.436,
+      "eval_samples_per_second": 196.742,
+      "eval_steps_per_second": 24.738,
+      "step": 3604
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 42400,
+  "num_train_epochs": 200,
+  "save_steps": 500,
+  "total_flos": 4842835489578468.0,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-3604/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11396e77b2185e0d94bce5383c2512c5afd7f340883723b782abc68f5d11c5a6
+size 4155