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:fa88b9923d32000ee2b4cdfc9e8d5b21092a959dfc406c3968dcb5b73db420a9
+size 544373707

checkpoint-120/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c7ee9f68ca0b9288e0b84b23d77fb474c898c21f60d253a0bbf4b2169d57cba
+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": 7.832551956176758,
+  "best_model_checkpoint": "/mmfs1/gscratch/stf/abhinavp/corpus-filtering/outputs/fict-full-lstm-42/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": 10.80423355102539,
+      "eval_runtime": 2.8085,
+      "eval_samples_per_second": 356.059,
+      "eval_steps_per_second": 44.507,
+      "step": 10
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.2e-05,
+      "loss": 10.7946,
+      "step": 20
+    },
+    {
+      "epoch": 0.16,
+      "eval_loss": 10.781017303466797,
+      "eval_runtime": 2.8659,
+      "eval_samples_per_second": 348.927,
+      "eval_steps_per_second": 43.616,
+      "step": 20
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 3.8e-05,
+      "loss": 10.7603,
+      "step": 30
+    },
+    {
+      "epoch": 0.24,
+      "eval_loss": 10.720499992370605,
+      "eval_runtime": 2.7648,
+      "eval_samples_per_second": 361.685,
+      "eval_steps_per_second": 45.211,
+      "step": 30
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.4000000000000007e-05,
+      "loss": 10.6159,
+      "step": 40
+    },
+    {
+      "epoch": 0.32,
+      "eval_loss": 10.284521102905273,
+      "eval_runtime": 2.8068,
+      "eval_samples_per_second": 356.277,
+      "eval_steps_per_second": 44.535,
+      "step": 40
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3e-05,
+      "loss": 9.881,
+      "step": 50
+    },
+    {
+      "epoch": 0.4,
+      "eval_loss": 9.241220474243164,
+      "eval_runtime": 2.8265,
+      "eval_samples_per_second": 353.795,
+      "eval_steps_per_second": 44.224,
+      "step": 50
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 2.6000000000000002e-05,
+      "loss": 9.0551,
+      "step": 60
+    },
+    {
+      "epoch": 0.48,
+      "eval_loss": 8.63198471069336,
+      "eval_runtime": 2.913,
+      "eval_samples_per_second": 343.289,
+      "eval_steps_per_second": 42.911,
+      "step": 60
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 2.2000000000000003e-05,
+      "loss": 8.5603,
+      "step": 70
+    },
+    {
+      "epoch": 0.56,
+      "eval_loss": 8.300483703613281,
+      "eval_runtime": 2.8122,
+      "eval_samples_per_second": 355.598,
+      "eval_steps_per_second": 44.45,
+      "step": 70
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8e-05,
+      "loss": 8.3761,
+      "step": 80
+    },
+    {
+      "epoch": 0.64,
+      "eval_loss": 8.105779647827148,
+      "eval_runtime": 2.8076,
+      "eval_samples_per_second": 356.172,
+      "eval_steps_per_second": 44.521,
+      "step": 80
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.4000000000000001e-05,
+      "loss": 8.2158,
+      "step": 90
+    },
+    {
+      "epoch": 0.72,
+      "eval_loss": 7.982827663421631,
+      "eval_runtime": 2.8051,
+      "eval_samples_per_second": 356.497,
+      "eval_steps_per_second": 44.562,
+      "step": 90
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1e-05,
+      "loss": 8.0922,
+      "step": 100
+    },
+    {
+      "epoch": 0.8,
+      "eval_loss": 7.904599189758301,
+      "eval_runtime": 2.8231,
+      "eval_samples_per_second": 354.225,
+      "eval_steps_per_second": 44.278,
+      "step": 100
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 6e-06,
+      "loss": 8.0326,
+      "step": 110
+    },
+    {
+      "epoch": 0.88,
+      "eval_loss": 7.856319427490234,
+      "eval_runtime": 2.8253,
+      "eval_samples_per_second": 353.939,
+      "eval_steps_per_second": 44.242,
+      "step": 110
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 8.1135,
+      "step": 120
+    },
+    {
+      "epoch": 0.96,
+      "eval_loss": 7.832551956176758,
+      "eval_runtime": 2.8376,
+      "eval_samples_per_second": 352.411,
+      "eval_steps_per_second": 44.051,
+      "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:ca94376b560130659aac7fc1141fac166f9e87d222e6f51cda86dacf6c253994
+size 4155