Training in progress, epoch 14, checkpoint

checkpoint-2030/config.json

@@ -0,0 +1,33 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 1,
+  "dtype": "float32",
+  "embd_pdrop": 0.1,
+  "eos_token_id": 2,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_ctx": 1024,
+  "n_embd": 768,
+  "n_head": 12,
+  "n_inner": null,
+  "n_layer": 12,
+  "n_positions": 1024,
+  "pad_token_id": 2,
+  "reorder_and_upcast_attn": false,
+  "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": false,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "transformers_version": "4.56.1",
+  "use_cache": true,
+  "vocab_size": 16000
+}

checkpoint-2030/generation_config.json

@@ -0,0 +1,9 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": [
+    2
+  ],
+  "pad_token_id": 2,
+  "transformers_version": "4.56.1"
+}

checkpoint-2030/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:267261d6926a79d770e2f9b249aaeb0747bed011a3b594227f34615c61f9ffd1
+size 392536704

checkpoint-2030/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18486770f32944aae452d520c300cb83e94b63d1c307029e0cccc7024d88a6d4
+size 785167627

checkpoint-2030/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b82b78b027ede29ee346c5118a0dbf9b1b714a12508a8c0f388380cbfe9d2fc8
+size 14709

checkpoint-2030/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2827dfd5ed4fb059f85622d433ac3be2dd8367200c576ad922e62884733a4475
+size 1383

checkpoint-2030/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ff40bccd031e61efec31894dc6d72a62020978e31d3a15d274981614efdc6ec
+size 1465

checkpoint-2030/special_tokens_map.json

@@ -0,0 +1,8 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "pad_token": "</s>",
+  "sep_token": "<sep>",
+  "unk_token": "<unk>"
+}

checkpoint-2030/tokenizer_config.json

@@ -0,0 +1,62 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "<sep>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "sep_token": "<sep>",
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "unk_token": "<unk>"
+}

checkpoint-2030/trainer_state.json

@@ -0,0 +1,251 @@
+{
+  "best_global_step": 1885,
+  "best_metric": 2.8720178604125977,
+  "best_model_checkpoint": "/home/p318482/syntactic-bootstrapping/model_trained/cds_o_fr_13/checkpoint-1885",
+  "epoch": 14.0,
+  "eval_steps": 500,
+  "global_step": 2030,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006896551724137931,
+      "grad_norm": 32.2354736328125,
+      "learning_rate": 0.0,
+      "loss": 9.7326,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.26012122631073,
+      "learning_rate": 2.88e-05,
+      "loss": 5.8188,
+      "step": 145
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 4.1781511306762695,
+      "eval_runtime": 4.3842,
+      "eval_samples_per_second": 803.562,
+      "eval_steps_per_second": 3.193,
+      "step": 145
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.0639448165893555,
+      "learning_rate": 5.7799999999999995e-05,
+      "loss": 3.6451,
+      "step": 290
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 3.4042274951934814,
+      "eval_runtime": 4.6134,
+      "eval_samples_per_second": 763.65,
+      "eval_steps_per_second": 3.035,
+      "step": 290
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.9253916144371033,
+      "learning_rate": 8.680000000000001e-05,
+      "loss": 3.2742,
+      "step": 435
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 3.1871676445007324,
+      "eval_runtime": 4.7952,
+      "eval_samples_per_second": 734.687,
+      "eval_steps_per_second": 2.92,
+      "step": 435
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.8254706263542175,
+      "learning_rate": 9.668067226890756e-05,
+      "loss": 3.1065,
+      "step": 580
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 3.0691733360290527,
+      "eval_runtime": 4.4368,
+      "eval_samples_per_second": 794.047,
+      "eval_steps_per_second": 3.155,
+      "step": 580
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.7847094535827637,
+      "learning_rate": 9.058823529411765e-05,
+      "loss": 3.0059,
+      "step": 725
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 3.00063157081604,
+      "eval_runtime": 4.5738,
+      "eval_samples_per_second": 770.265,
+      "eval_steps_per_second": 3.061,
+      "step": 725
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.82077956199646,
+      "learning_rate": 8.449579831932774e-05,
+      "loss": 2.9407,
+      "step": 870
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.95955753326416,
+      "eval_runtime": 4.664,
+      "eval_samples_per_second": 755.368,
+      "eval_steps_per_second": 3.002,
+      "step": 870
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 0.7905252575874329,
+      "learning_rate": 7.840336134453782e-05,
+      "loss": 2.8903,
+      "step": 1015
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 2.929441452026367,
+      "eval_runtime": 4.4917,
+      "eval_samples_per_second": 784.335,
+      "eval_steps_per_second": 3.117,
+      "step": 1015
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 0.8253563642501831,
+      "learning_rate": 7.23109243697479e-05,
+      "loss": 2.8492,
+      "step": 1160
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 2.907557487487793,
+      "eval_runtime": 4.4334,
+      "eval_samples_per_second": 794.644,
+      "eval_steps_per_second": 3.158,
+      "step": 1160
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 0.8737186789512634,
+      "learning_rate": 6.621848739495798e-05,
+      "loss": 2.8124,
+      "step": 1305
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 2.894392728805542,
+      "eval_runtime": 4.3693,
+      "eval_samples_per_second": 806.31,
+      "eval_steps_per_second": 3.204,
+      "step": 1305
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 0.9210653901100159,
+      "learning_rate": 6.012605042016807e-05,
+      "loss": 2.7787,
+      "step": 1450
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 2.8836522102355957,
+      "eval_runtime": 4.3707,
+      "eval_samples_per_second": 806.047,
+      "eval_steps_per_second": 3.203,
+      "step": 1450
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 0.9565634727478027,
+      "learning_rate": 5.403361344537815e-05,
+      "loss": 2.7466,
+      "step": 1595
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 2.8761401176452637,
+      "eval_runtime": 4.46,
+      "eval_samples_per_second": 789.914,
+      "eval_steps_per_second": 3.139,
+      "step": 1595
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.0066618919372559,
+      "learning_rate": 4.794117647058824e-05,
+      "loss": 2.7144,
+      "step": 1740
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 2.8736512660980225,
+      "eval_runtime": 4.4277,
+      "eval_samples_per_second": 795.673,
+      "eval_steps_per_second": 3.162,
+      "step": 1740
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.0534671545028687,
+      "learning_rate": 4.184873949579832e-05,
+      "loss": 2.6832,
+      "step": 1885
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 2.8720178604125977,
+      "eval_runtime": 4.4634,
+      "eval_samples_per_second": 789.313,
+      "eval_steps_per_second": 3.137,
+      "step": 1885
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.1330785751342773,
+      "learning_rate": 3.575630252100841e-05,
+      "loss": 2.6516,
+      "step": 2030
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 2.874244451522827,
+      "eval_runtime": 4.764,
+      "eval_samples_per_second": 739.504,
+      "eval_steps_per_second": 2.939,
+      "step": 2030
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 2900,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 20,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.3822685790208e+16,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-2030/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35d1173395a906c65a769a2df8203d374e82540a954fe0dc718fba07dd3acdc2
+size 5905