Training in progress, epoch 1

emissions.csv

@@ -5,4 +5,5 @@ timestamp,project_name,run_id,duration,emissions,emissions_rate,cpu_power,gpu_po
 2024-03-20T11:07:57,codecarbon,8c5cd6be-b949-4d3e-aeee-44d3635c6e6a,129.25563645362854,0.0002480293691227,1.918905634817298e-06,42.5,100.54736914935307,96.0,0.0015258966469102,0.0040307021134466,0.0034462821515401,0.009002880911897,Norway,NOR,viken,,,Linux-5.14.0-362.8.1.el9_3.x86_64-x86_64-with-glibc2.34,3.11.3,2.3.4,24,AMD EPYC 7642 48-Core Processor,1,1 x NVIDIA A100-PCIE-40GB,10.8034,59.6573,256,machine,N,1.0
 2024-03-20T11:09:11,codecarbon,8c5cd6be-b949-4d3e-aeee-44d3635c6e6a,202.64030289649963,0.000394073813688,1.944696134259563e-06,42.5,121.5938351011336,96.0,0.002392233763304,0.006508886873771,0.0054028290430704,0.0143039496801455,Norway,NOR,viken,,,Linux-5.14.0-362.8.1.el9_3.x86_64-x86_64-with-glibc2.34,3.11.3,2.3.4,24,AMD EPYC 7642 48-Core Processor,1,1 x NVIDIA A100-PCIE-40GB,10.8034,59.6573,256,machine,N,1.0
 2024-03-20T11:10:18,codecarbon,8c5cd6be-b949-4d3e-aeee-44d3635c6e6a,269.8455903530121,0.0005163006217623,1.913318728265911e-06,42.5,99.17360382771624,96.0,0.003185620978309,0.008360091132511,0.0071947823270161,0.0187404944378362,Norway,NOR,viken,,,Linux-5.14.0-362.8.1.el9_3.x86_64-x86_64-with-glibc2.34,3.11.3,2.3.4,24,AMD EPYC 7642 48-Core Processor,1,1 x NVIDIA A100-PCIE-40GB,10.8034,59.6573,256,machine,N,1.0
-2024-03-20T11:14:00,codecarbon,8c5cd6be-b949-4d3e-aeee-44d3635c6e6a,492.169837474823,0.000945341925447641,1.920763633744622e-06,42.5,113.67679084499753,96.0,0.0058102714010410846,0.015380181470801446,0.01312322863260905,0.034313681504451576,Norway,NOR,viken,,,Linux-5.14.0-362.8.1.el9_3.x86_64-x86_64-with-glibc2.34,3.11.3,2.3.4,24,AMD EPYC 7642 48-Core Processor,1,1 x NVIDIA A100-PCIE-40GB,10.8034,59.6573,256,machine,N,1.0
+2024-03-20T11:14:00,codecarbon,8c5cd6be-b949-4d3e-aeee-44d3635c6e6a,492.169837474823,0.0009453419254476,1.920763633744622e-06,42.5,113.67679084499753,96.0,0.005810271401041,0.0153801814708014,0.013123228632609,0.0343136815044515,Norway,NOR,viken,,,Linux-5.14.0-362.8.1.el9_3.x86_64-x86_64-with-glibc2.34,3.11.3,2.3.4,24,AMD EPYC 7642 48-Core Processor,1,1 x NVIDIA A100-PCIE-40GB,10.8034,59.6573,256,machine,N,1.0
+2024-03-20T11:15:05,codecarbon,8c5cd6be-b949-4d3e-aeee-44d3635c6e6a,556.7771844863892,0.001067438719091254,1.917173959051385e-06,42.5,108.4736124826506,96.0,0.006572990199923516,0.017326653305753226,0.014845863539377849,0.03874550704505459,Norway,NOR,viken,,,Linux-5.14.0-362.8.1.el9_3.x86_64-x86_64-with-glibc2.34,3.11.3,2.3.4,24,AMD EPYC 7642 48-Core Processor,1,1 x NVIDIA A100-PCIE-40GB,10.8034,59.6573,256,machine,N,1.0

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4df5f123a2571aafa582671c239fee4b33e10950d1fe7e12283e07f3b26c89d8
+oid sha256:243a6355c9ccdf996b8c52460c68b2382106a7e923e7402167b0e7a83fdce44c
 size 267832560

run-5/checkpoint-134/config.json

@@ -0,0 +1,25 @@
+{
+  "_name_or_path": "distilbert-base-uncased",
+  "activation": "gelu",
+  "architectures": [
+    "DistilBertForSequenceClassification"
+  ],
+  "attention_dropout": 0.1,
+  "dim": 768,
+  "dropout": 0.1,
+  "hidden_dim": 3072,
+  "initializer_range": 0.02,
+  "max_position_embeddings": 512,
+  "model_type": "distilbert",
+  "n_heads": 12,
+  "n_layers": 6,
+  "pad_token_id": 0,
+  "problem_type": "single_label_classification",
+  "qa_dropout": 0.1,
+  "seq_classif_dropout": 0.2,
+  "sinusoidal_pos_embds": false,
+  "tie_weights_": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.38.2",
+  "vocab_size": 30522
+}

run-5/checkpoint-134/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:243a6355c9ccdf996b8c52460c68b2382106a7e923e7402167b0e7a83fdce44c
+size 267832560

run-5/checkpoint-134/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c91b97f7a56dbfb043fa1868dabe922e3a7670b1c318987ec70603ede404809
+size 535727290

run-5/checkpoint-134/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9212ee7fb09c4452011b2d10e24f721aae3807afdd0832bede26eeae82b4712b
+size 14244

run-5/checkpoint-134/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7d6dbb0133016f4d0ef63947f78694ab4651260b0d15d6bd4f587fcf0b5aa8d
+size 1064

run-5/checkpoint-134/special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

run-5/checkpoint-134/tokenizer_config.json

@@ -0,0 +1,55 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "DistilBertTokenizer",
+  "unk_token": "[UNK]"
+}

run-5/checkpoint-134/trainer_state.json

@@ -0,0 +1,35 @@
+{
+  "best_metric": 0.3429695650358,
+  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-5/checkpoint-134",
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 134,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.5264953970909119,
+      "eval_matthews_correlation": 0.3429695650358,
+      "eval_runtime": 0.34,
+      "eval_samples_per_second": 3068.067,
+      "eval_steps_per_second": 194.144,
+      "step": 134
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 536,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 500,
+  "total_flos": 0,
+  "train_batch_size": 64,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 1.1390296624220138e-05,
+    "num_train_epochs": 4,
+    "per_device_train_batch_size": 64,
+    "seed": 11
+  }
+}

run-5/checkpoint-134/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b0a20b7e1ced1e066ae0be151da2892022f2d3caa0764af95b56a89f7778c56
+size 4984

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e09d686d744a0008ff792f3863883364cd703287c8e9332ee8d8209d508c0a5d
+oid sha256:7b0a20b7e1ced1e066ae0be151da2892022f2d3caa0764af95b56a89f7778c56
 size 4984