Training in progress, epoch 13, checkpoint

checkpoint-1885/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-1885/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-1885/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7039d3b7b640004147395487d83ccb824459874edb47b950cfad980c004f8aa4
+size 392536704

checkpoint-1885/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21c52aebad7425939f63a3592fad8de2cd07eb0d9590a9f72190224ac48edd56
+size 785167627

checkpoint-1885/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:958e45fe14e50ebde759ff138d0274fc79cc41ab11bbc8eb602f431800179ba7
+size 14709

checkpoint-1885/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1863ededb286660ffbe645acd90efcdef4f55745af6b660ebe9fd977f69c8051
+size 1383

checkpoint-1885/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:65f4171fd385abefff02949fc64463190865751b481933fdf0de467e1ff48fb4
+size 1465

checkpoint-1885/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-1885/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-1885/trainer_state.json

@@ -0,0 +1,236 @@
+{
+  "best_global_step": 1885,
+  "best_metric": 3.294769048690796,
+  "best_model_checkpoint": "/home/p318482/syntactic-bootstrapping/model_trained/cds_sh1_fr_13/checkpoint-1885",
+  "epoch": 13.0,
+  "eval_steps": 500,
+  "global_step": 1885,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006896551724137931,
+      "grad_norm": 32.49538040161133,
+      "learning_rate": 0.0,
+      "loss": 9.7326,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.3874874114990234,
+      "learning_rate": 2.88e-05,
+      "loss": 6.0116,
+      "step": 145
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 4.523594856262207,
+      "eval_runtime": 4.3671,
+      "eval_samples_per_second": 806.248,
+      "eval_steps_per_second": 3.206,
+      "step": 145
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.230023741722107,
+      "learning_rate": 5.7799999999999995e-05,
+      "loss": 4.17,
+      "step": 290
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 4.019691467285156,
+      "eval_runtime": 4.7428,
+      "eval_samples_per_second": 742.384,
+      "eval_steps_per_second": 2.952,
+      "step": 290
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.5065193176269531,
+      "learning_rate": 8.680000000000001e-05,
+      "loss": 3.8789,
+      "step": 435
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 3.7656915187835693,
+      "eval_runtime": 4.7079,
+      "eval_samples_per_second": 747.885,
+      "eval_steps_per_second": 2.974,
+      "step": 435
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.2275551557540894,
+      "learning_rate": 9.668067226890756e-05,
+      "loss": 3.6772,
+      "step": 580
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 3.6136202812194824,
+      "eval_runtime": 4.4571,
+      "eval_samples_per_second": 789.983,
+      "eval_steps_per_second": 3.141,
+      "step": 580
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.2071841955184937,
+      "learning_rate": 9.058823529411765e-05,
+      "loss": 3.543,
+      "step": 725
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 3.5114729404449463,
+      "eval_runtime": 4.4391,
+      "eval_samples_per_second": 793.177,
+      "eval_steps_per_second": 3.154,
+      "step": 725
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.2674086093902588,
+      "learning_rate": 8.449579831932774e-05,
+      "loss": 3.452,
+      "step": 870
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 3.456989049911499,
+      "eval_runtime": 4.3824,
+      "eval_samples_per_second": 803.444,
+      "eval_steps_per_second": 3.195,
+      "step": 870
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.297289252281189,
+      "learning_rate": 7.840336134453782e-05,
+      "loss": 3.3816,
+      "step": 1015
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 3.412864923477173,
+      "eval_runtime": 4.4639,
+      "eval_samples_per_second": 788.768,
+      "eval_steps_per_second": 3.136,
+      "step": 1015
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.2838830947875977,
+      "learning_rate": 7.23109243697479e-05,
+      "loss": 3.3218,
+      "step": 1160
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 3.370572805404663,
+      "eval_runtime": 4.3865,
+      "eval_samples_per_second": 802.694,
+      "eval_steps_per_second": 3.192,
+      "step": 1160
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 1.3836426734924316,
+      "learning_rate": 6.621848739495798e-05,
+      "loss": 3.2718,
+      "step": 1305
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 3.3445897102355957,
+      "eval_runtime": 4.3621,
+      "eval_samples_per_second": 807.184,
+      "eval_steps_per_second": 3.209,
+      "step": 1305
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.383060336112976,
+      "learning_rate": 6.012605042016807e-05,
+      "loss": 3.2257,
+      "step": 1450
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 3.3246004581451416,
+      "eval_runtime": 4.3747,
+      "eval_samples_per_second": 804.851,
+      "eval_steps_per_second": 3.2,
+      "step": 1450
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.4625455141067505,
+      "learning_rate": 5.403361344537815e-05,
+      "loss": 3.183,
+      "step": 1595
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 3.3089232444763184,
+      "eval_runtime": 4.3848,
+      "eval_samples_per_second": 803.01,
+      "eval_steps_per_second": 3.193,
+      "step": 1595
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.4933857917785645,
+      "learning_rate": 4.794117647058824e-05,
+      "loss": 3.1406,
+      "step": 1740
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 3.302008867263794,
+      "eval_runtime": 4.3752,
+      "eval_samples_per_second": 804.771,
+      "eval_steps_per_second": 3.2,
+      "step": 1740
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.5790159702301025,
+      "learning_rate": 4.184873949579832e-05,
+      "loss": 3.1026,
+      "step": 1885
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 3.294769048690796,
+      "eval_runtime": 4.3716,
+      "eval_samples_per_second": 805.427,
+      "eval_steps_per_second": 3.202,
+      "step": 1885
+    }
+  ],
+  "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": 3.1361772109824e+16,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-1885/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2511e72b4eb6d24894b9928038d96187a2b1678f794de3fab248baf85c0fa210
+size 5905