Training in progress, step 70, checkpoint

checkpoint-70/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-70/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:280d036a5127241d21174b54dd3f198d65b0b4a23227958899012bc9e604bcde
+size 544373707

checkpoint-70/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ccc62a82fbf2bfa183e42488ddbb6388249f46b596772d585c9286898c1df67
+size 272184705

checkpoint-70/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72f9c5cb78d327d1ddf367de793b66155aa4f2e9eba4c3b94484ff71b3bd6c4c
+size 14575

checkpoint-70/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b4a4f2c85b94c163137ac0f7e7f7d58f8c179849484499985cc479a84c6cba40
+size 627

checkpoint-70/special_tokens_map.json

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

checkpoint-70/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-70/trainer_state.json

@@ -0,0 +1,117 @@
+{
+  "best_metric": 8.260101318359375,
+  "best_model_checkpoint": "/mmfs1/gscratch/stf/abhinavp/corpus-filtering/outputs/2023-11-20/16-19-50/checkpoints/checkpoint-70",
+  "epoch": 0.56,
+  "eval_steps": 10,
+  "global_step": 70,
+  "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": 10.8117,
+      "step": 10
+    },
+    {
+      "epoch": 0.08,
+      "eval_loss": 10.801591873168945,
+      "eval_runtime": 2.6865,
+      "eval_samples_per_second": 372.226,
+      "eval_steps_per_second": 46.528,
+      "step": 10
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.2e-05,
+      "loss": 10.7909,
+      "step": 20
+    },
+    {
+      "epoch": 0.16,
+      "eval_loss": 10.775726318359375,
+      "eval_runtime": 2.779,
+      "eval_samples_per_second": 359.838,
+      "eval_steps_per_second": 44.98,
+      "step": 20
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 3.8e-05,
+      "loss": 10.7524,
+      "step": 30
+    },
+    {
+      "epoch": 0.24,
+      "eval_loss": 10.704446792602539,
+      "eval_runtime": 2.7821,
+      "eval_samples_per_second": 359.435,
+      "eval_steps_per_second": 44.929,
+      "step": 30
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.4000000000000007e-05,
+      "loss": 10.5629,
+      "step": 40
+    },
+    {
+      "epoch": 0.32,
+      "eval_loss": 10.172320365905762,
+      "eval_runtime": 2.7542,
+      "eval_samples_per_second": 363.084,
+      "eval_steps_per_second": 45.385,
+      "step": 40
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3e-05,
+      "loss": 9.7608,
+      "step": 50
+    },
+    {
+      "epoch": 0.4,
+      "eval_loss": 9.128562927246094,
+      "eval_runtime": 2.7812,
+      "eval_samples_per_second": 359.553,
+      "eval_steps_per_second": 44.944,
+      "step": 50
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 2.6000000000000002e-05,
+      "loss": 8.9735,
+      "step": 60
+    },
+    {
+      "epoch": 0.48,
+      "eval_loss": 8.557924270629883,
+      "eval_runtime": 2.6727,
+      "eval_samples_per_second": 374.147,
+      "eval_steps_per_second": 46.768,
+      "step": 60
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 2.2000000000000003e-05,
+      "loss": 8.5033,
+      "step": 70
+    },
+    {
+      "epoch": 0.56,
+      "eval_loss": 8.260101318359375,
+      "eval_runtime": 2.7328,
+      "eval_samples_per_second": 365.926,
+      "eval_steps_per_second": 45.741,
+      "step": 70
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 125,
+  "num_train_epochs": 1,
+  "save_steps": 10,
+  "total_flos": 4923721736640.0,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-70/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3bffe3ee8fa6f55af146c4d5f06cfb3e6ce7d7e37047e3711004138250f71a94
+size 4155