Training in progress, step 120, checkpoint

checkpoint-120/config.json

@@ -0,0 +1,24 @@
+{
+  "architectures": [
+    "RNNForLanguageModeling"
+  ],
+  "bidirectional": false,
+  "dropout_p": 0.1,
+  "emb_init_range": 0.1,
+  "embedding_dim": 1024,
+  "embedding_kwargs": {},
+  "hidden_dim": 1024,
+  "lin_init_range": 0.03125,
+  "lm_in_features": 1024,
+  "model_type": "rnn",
+  "num_layers": 2,
+  "output_last_state": false,
+  "output_recurrent_outputs": false,
+  "recur_init_range": 0.03125,
+  "rnn_kwargs": {},
+  "rnn_type": "LSTM",
+  "tie_weights": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.33.3",
+  "vocab_size": 50002
+}

checkpoint-120/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:78aeea20a8db3690cd249c6a823fcef55e8b3167ca8a0d31af65bbb3618004e9
+size 544373707

checkpoint-120/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b063ac7153ced1938972acd4df28a12a4f0ee7aab68af526d0674d22c9e74873
+size 272184705

checkpoint-120/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ace41d3c0f4cbdbd37c8c82ec4459c32155e7fd93c6d29595070e0d356a90d0f
+size 14575

checkpoint-120/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83733202736c46393c86334f4c75823e9b1535759987f1877b9903c178ce0959
+size 627

checkpoint-120/special_tokens_map.json

@@ -0,0 +1,5 @@
+{
+  "eos_token": "<eos>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

checkpoint-120/tokenizer_config.json

@@ -0,0 +1,12 @@
+{
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<eos>",
+  "max_length": null,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_to_multiple_of": null,
+  "pad_token": "<pad>",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "unk_token": "<unk>"
+}

checkpoint-120/trainer_state.json

@@ -0,0 +1,187 @@
+{
+  "best_metric": 7.829923629760742,
+  "best_model_checkpoint": "/mmfs1/gscratch/stf/abhinavp/corpus-filtering/outputs/2023-11-20/14-54-26/checkpoints/checkpoint-120",
+  "epoch": 0.96,
+  "eval_steps": 10,
+  "global_step": 120,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 10.8138,
+      "step": 10
+    },
+    {
+      "epoch": 0.08,
+      "eval_loss": 10.804353713989258,
+      "eval_runtime": 2.6966,
+      "eval_samples_per_second": 370.843,
+      "eval_steps_per_second": 46.355,
+      "step": 10
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.2e-05,
+      "loss": 10.7944,
+      "step": 20
+    },
+    {
+      "epoch": 0.16,
+      "eval_loss": 10.780308723449707,
+      "eval_runtime": 2.9168,
+      "eval_samples_per_second": 342.842,
+      "eval_steps_per_second": 42.855,
+      "step": 20
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 3.8e-05,
+      "loss": 10.7582,
+      "step": 30
+    },
+    {
+      "epoch": 0.24,
+      "eval_loss": 10.715201377868652,
+      "eval_runtime": 2.6781,
+      "eval_samples_per_second": 373.4,
+      "eval_steps_per_second": 46.675,
+      "step": 30
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.4000000000000007e-05,
+      "loss": 10.5902,
+      "step": 40
+    },
+    {
+      "epoch": 0.32,
+      "eval_loss": 10.216275215148926,
+      "eval_runtime": 2.8449,
+      "eval_samples_per_second": 351.507,
+      "eval_steps_per_second": 43.938,
+      "step": 40
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3e-05,
+      "loss": 9.8089,
+      "step": 50
+    },
+    {
+      "epoch": 0.4,
+      "eval_loss": 9.195159912109375,
+      "eval_runtime": 2.6916,
+      "eval_samples_per_second": 371.531,
+      "eval_steps_per_second": 46.441,
+      "step": 50
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 2.6000000000000002e-05,
+      "loss": 9.0218,
+      "step": 60
+    },
+    {
+      "epoch": 0.48,
+      "eval_loss": 8.602910041809082,
+      "eval_runtime": 2.6711,
+      "eval_samples_per_second": 374.379,
+      "eval_steps_per_second": 46.797,
+      "step": 60
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 2.2000000000000003e-05,
+      "loss": 8.5416,
+      "step": 70
+    },
+    {
+      "epoch": 0.56,
+      "eval_loss": 8.279220581054688,
+      "eval_runtime": 2.7623,
+      "eval_samples_per_second": 362.019,
+      "eval_steps_per_second": 45.252,
+      "step": 70
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8e-05,
+      "loss": 8.3655,
+      "step": 80
+    },
+    {
+      "epoch": 0.64,
+      "eval_loss": 8.092720985412598,
+      "eval_runtime": 2.7605,
+      "eval_samples_per_second": 362.25,
+      "eval_steps_per_second": 45.281,
+      "step": 80
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.4000000000000001e-05,
+      "loss": 8.2139,
+      "step": 90
+    },
+    {
+      "epoch": 0.72,
+      "eval_loss": 7.974403381347656,
+      "eval_runtime": 2.6732,
+      "eval_samples_per_second": 374.09,
+      "eval_steps_per_second": 46.761,
+      "step": 90
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 8.1098,
+      "step": 100
+    },
+    {
+      "epoch": 0.8,
+      "eval_loss": 7.899349689483643,
+      "eval_runtime": 2.8812,
+      "eval_samples_per_second": 347.075,
+      "eval_steps_per_second": 43.384,
+      "step": 100
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 6e-06,
+      "loss": 8.0232,
+      "step": 110
+    },
+    {
+      "epoch": 0.88,
+      "eval_loss": 7.852936267852783,
+      "eval_runtime": 2.6664,
+      "eval_samples_per_second": 375.035,
+      "eval_steps_per_second": 46.879,
+      "step": 110
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 8.1088,
+      "step": 120
+    },
+    {
+      "epoch": 0.96,
+      "eval_loss": 7.829923629760742,
+      "eval_runtime": 2.8976,
+      "eval_samples_per_second": 345.118,
+      "eval_steps_per_second": 43.14,
+      "step": 120
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 125,
+  "num_train_epochs": 1,
+  "save_steps": 10,
+  "total_flos": 8440665834240.0,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-120/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63ae9edac8e41c0240cbb09648ef4171e2ce67c1c9132186453148fa757a3f68
+size 4155