Training in progress, epoch 17, checkpoint

checkpoint-1853/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-1853/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-1853/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7268d25cca6758efe786eeb39f9aea89a900898988aedc0e3f8ac043c86b8a16
+size 392536704

checkpoint-1853/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77e498f409aedd0be461a197d2538412532f512d0c244f278cb2f62cc518947c
+size 785167627

checkpoint-1853/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1c22a7755d523ce8ce154d3725ae56dde3fb41f309e46a69ebf4e6bd4f13bfa
+size 14709

checkpoint-1853/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24a67c93d6704c3099268705593e2b7417f8260dfc4cb137610bee590fc495c6
+size 1383

checkpoint-1853/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c8e68f7749acf06f4e3f5acd1cd077fbb1d2765fe58ea2e4917a40ce91dea03
+size 1465

checkpoint-1853/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-1853/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-1853/trainer_state.json

@@ -0,0 +1,296 @@
+{
+  "best_global_step": 1635,
+  "best_metric": 3.888904333114624,
+  "best_model_checkpoint": "/home/p318482/syntactic-bootstrapping/model_trained/ads_o_fr_13/checkpoint-1635",
+  "epoch": 17.0,
+  "eval_steps": 500,
+  "global_step": 1853,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.009174311926605505,
+      "grad_norm": 17.951679229736328,
+      "learning_rate": 0.0,
+      "loss": 9.7117,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 2.537048101425171,
+      "learning_rate": 2.16e-05,
+      "loss": 7.4816,
+      "step": 109
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 6.125095367431641,
+      "eval_runtime": 3.3374,
+      "eval_samples_per_second": 810.812,
+      "eval_steps_per_second": 3.296,
+      "step": 109
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.5311152935028076,
+      "learning_rate": 4.3400000000000005e-05,
+      "loss": 5.2479,
+      "step": 218
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 4.817795753479004,
+      "eval_runtime": 3.4684,
+      "eval_samples_per_second": 780.181,
+      "eval_steps_per_second": 3.171,
+      "step": 218
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.4448567628860474,
+      "learning_rate": 6.52e-05,
+      "loss": 4.5633,
+      "step": 327
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 4.496043682098389,
+      "eval_runtime": 3.3595,
+      "eval_samples_per_second": 805.47,
+      "eval_steps_per_second": 3.274,
+      "step": 327
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.4543994665145874,
+      "learning_rate": 8.7e-05,
+      "loss": 4.3166,
+      "step": 436
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 4.303887844085693,
+      "eval_runtime": 3.3467,
+      "eval_samples_per_second": 808.547,
+      "eval_steps_per_second": 3.287,
+      "step": 436
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.3640012741088867,
+      "learning_rate": 9.734939759036145e-05,
+      "loss": 4.1561,
+      "step": 545
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 4.18064546585083,
+      "eval_runtime": 3.442,
+      "eval_samples_per_second": 786.172,
+      "eval_steps_per_second": 3.196,
+      "step": 545
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.4212088584899902,
+      "learning_rate": 9.078313253012048e-05,
+      "loss": 4.0364,
+      "step": 654
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 4.096184253692627,
+      "eval_runtime": 3.3717,
+      "eval_samples_per_second": 802.573,
+      "eval_steps_per_second": 3.262,
+      "step": 654
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.3861104249954224,
+      "learning_rate": 8.421686746987952e-05,
+      "loss": 3.9474,
+      "step": 763
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 4.03432559967041,
+      "eval_runtime": 3.3677,
+      "eval_samples_per_second": 803.523,
+      "eval_steps_per_second": 3.266,
+      "step": 763
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.4673441648483276,
+      "learning_rate": 7.765060240963856e-05,
+      "loss": 3.8753,
+      "step": 872
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 3.990006685256958,
+      "eval_runtime": 3.3655,
+      "eval_samples_per_second": 804.045,
+      "eval_steps_per_second": 3.268,
+      "step": 872
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 1.5565500259399414,
+      "learning_rate": 7.108433734939759e-05,
+      "loss": 3.8118,
+      "step": 981
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 3.9578440189361572,
+      "eval_runtime": 3.351,
+      "eval_samples_per_second": 807.532,
+      "eval_steps_per_second": 3.283,
+      "step": 981
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.5941141843795776,
+      "learning_rate": 6.451807228915663e-05,
+      "loss": 3.7548,
+      "step": 1090
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 3.9341025352478027,
+      "eval_runtime": 3.5715,
+      "eval_samples_per_second": 757.67,
+      "eval_steps_per_second": 3.08,
+      "step": 1090
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.778639793395996,
+      "learning_rate": 5.7951807228915664e-05,
+      "loss": 3.7009,
+      "step": 1199
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 3.916402578353882,
+      "eval_runtime": 3.3516,
+      "eval_samples_per_second": 807.368,
+      "eval_steps_per_second": 3.282,
+      "step": 1199
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.8095911741256714,
+      "learning_rate": 5.1385542168674705e-05,
+      "loss": 3.6487,
+      "step": 1308
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 3.9003710746765137,
+      "eval_runtime": 3.6596,
+      "eval_samples_per_second": 739.417,
+      "eval_steps_per_second": 3.006,
+      "step": 1308
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 2.016235113143921,
+      "learning_rate": 4.481927710843374e-05,
+      "loss": 3.6004,
+      "step": 1417
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 3.892484188079834,
+      "eval_runtime": 3.3724,
+      "eval_samples_per_second": 802.396,
+      "eval_steps_per_second": 3.262,
+      "step": 1417
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 2.106367826461792,
+      "learning_rate": 3.8253012048192774e-05,
+      "loss": 3.5523,
+      "step": 1526
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 3.894266366958618,
+      "eval_runtime": 3.5059,
+      "eval_samples_per_second": 771.833,
+      "eval_steps_per_second": 3.138,
+      "step": 1526
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 2.2525172233581543,
+      "learning_rate": 3.168674698795181e-05,
+      "loss": 3.5049,
+      "step": 1635
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 3.888904333114624,
+      "eval_runtime": 3.51,
+      "eval_samples_per_second": 770.943,
+      "eval_steps_per_second": 3.134,
+      "step": 1635
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 2.453017473220825,
+      "learning_rate": 2.5120481927710842e-05,
+      "loss": 3.4614,
+      "step": 1744
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 3.8933465480804443,
+      "eval_runtime": 3.3572,
+      "eval_samples_per_second": 806.026,
+      "eval_steps_per_second": 3.277,
+      "step": 1744
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 2.57418155670166,
+      "learning_rate": 1.855421686746988e-05,
+      "loss": 3.4222,
+      "step": 1853
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 3.897146701812744,
+      "eval_runtime": 3.6657,
+      "eval_samples_per_second": 738.188,
+      "eval_steps_per_second": 3.001,
+      "step": 1853
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 2180,
+  "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": 3.0758383484928e+16,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-1853/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07d93deb681371270a2b11ffb08b575a730e512a2a8d3e5c01548c44e1aa4ff7
+size 5905