Training in progress, epoch 12, checkpoint

checkpoint-5952/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": 30000
+}

checkpoint-5952/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-5952/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40f7861a4bcb4ae97be4a78f36226206ee663b939f555bb9d3175ea6dd5244d7
+size 435544704

checkpoint-5952/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7693c49f8e826f9e5cadf64ab7eeead4d252fa50bc3a95dc8b1656b7feff061
+size 871183627

checkpoint-5952/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:798ada0b5eaf83acfdd5d70b3b86b1709fa917df9f1af38ace2084c855178926
+size 14709

checkpoint-5952/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ecd99389d0b0fa98b257c145b8eae39d68db39de8045d8bf547a8b89832a7a5
+size 1383

checkpoint-5952/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b164cd6220f6084c85fe49df30196c3d608786f72c95d3910b71b06fd18bb78a
+size 1465

checkpoint-5952/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-5952/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-5952/trainer_state.json

@@ -0,0 +1,221 @@
+{
+  "best_global_step": 4464,
+  "best_metric": 3.096350908279419,
+  "best_model_checkpoint": "/home/p318482/syntactic-bootstrapping/model_trained/cds_w_13/checkpoint-4464",
+  "epoch": 12.0,
+  "eval_steps": 500,
+  "global_step": 5952,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0020161290322580645,
+      "grad_norm": 30.076793670654297,
+      "learning_rate": 0.0,
+      "loss": 10.3404,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 4.318020343780518,
+      "learning_rate": 9.900000000000001e-05,
+      "loss": 4.5857,
+      "step": 496
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 3.524101972579956,
+      "eval_runtime": 43.6446,
+      "eval_samples_per_second": 729.46,
+      "eval_steps_per_second": 2.864,
+      "step": 496
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 4.235888481140137,
+      "learning_rate": 9.477659574468086e-05,
+      "loss": 3.3228,
+      "step": 992
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 3.3001248836517334,
+      "eval_runtime": 44.2288,
+      "eval_samples_per_second": 719.826,
+      "eval_steps_per_second": 2.826,
+      "step": 992
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 4.020275115966797,
+      "learning_rate": 8.950000000000001e-05,
+      "loss": 3.1656,
+      "step": 1488
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 3.2202703952789307,
+      "eval_runtime": 44.1563,
+      "eval_samples_per_second": 721.006,
+      "eval_steps_per_second": 2.831,
+      "step": 1488
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 4.647115707397461,
+      "learning_rate": 8.422340425531914e-05,
+      "loss": 3.078,
+      "step": 1984
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 3.1739180088043213,
+      "eval_runtime": 44.5914,
+      "eval_samples_per_second": 713.972,
+      "eval_steps_per_second": 2.803,
+      "step": 1984
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 3.8257627487182617,
+      "learning_rate": 7.894680851063831e-05,
+      "loss": 3.0125,
+      "step": 2480
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 3.143359661102295,
+      "eval_runtime": 43.8817,
+      "eval_samples_per_second": 725.518,
+      "eval_steps_per_second": 2.849,
+      "step": 2480
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 4.213743686676025,
+      "learning_rate": 7.367021276595744e-05,
+      "loss": 2.9601,
+      "step": 2976
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 3.1195075511932373,
+      "eval_runtime": 43.9406,
+      "eval_samples_per_second": 724.547,
+      "eval_steps_per_second": 2.845,
+      "step": 2976
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 4.123694896697998,
+      "learning_rate": 6.83936170212766e-05,
+      "loss": 2.9126,
+      "step": 3472
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 3.1072616577148438,
+      "eval_runtime": 43.9701,
+      "eval_samples_per_second": 724.061,
+      "eval_steps_per_second": 2.843,
+      "step": 3472
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 4.427916049957275,
+      "learning_rate": 6.311702127659575e-05,
+      "loss": 2.868,
+      "step": 3968
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 3.1035728454589844,
+      "eval_runtime": 44.0586,
+      "eval_samples_per_second": 722.605,
+      "eval_steps_per_second": 2.837,
+      "step": 3968
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 4.914698123931885,
+      "learning_rate": 5.7840425531914896e-05,
+      "loss": 2.8271,
+      "step": 4464
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 3.096350908279419,
+      "eval_runtime": 44.156,
+      "eval_samples_per_second": 721.011,
+      "eval_steps_per_second": 2.831,
+      "step": 4464
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 5.213223457336426,
+      "learning_rate": 5.256382978723404e-05,
+      "loss": 2.787,
+      "step": 4960
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 3.09796142578125,
+      "eval_runtime": 44.0364,
+      "eval_samples_per_second": 722.971,
+      "eval_steps_per_second": 2.839,
+      "step": 4960
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 6.033893585205078,
+      "learning_rate": 4.728723404255319e-05,
+      "loss": 2.7475,
+      "step": 5456
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 3.0985019207000732,
+      "eval_runtime": 43.99,
+      "eval_samples_per_second": 723.733,
+      "eval_steps_per_second": 2.842,
+      "step": 5456
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 6.661558628082275,
+      "learning_rate": 4.201063829787234e-05,
+      "loss": 2.71,
+      "step": 5952
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 3.1065890789031982,
+      "eval_runtime": 44.2679,
+      "eval_samples_per_second": 719.189,
+      "eval_steps_per_second": 2.824,
+      "step": 5952
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 9920,
+  "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": 9.9335914389504e+16,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-5952/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9ce9c5ddd61da8e500a622d9923d4f34a9d7d2dd637de93ee966a57021e3430
+size 5905