Training in progress, step 18000, checkpoint

checkpoint-18000/config.json

@@ -0,0 +1,32 @@
+{
+  "architectures": [
+    "LongelmForMaskedLMWithEdgePrediction"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "block_size": 8,
+  "bos_token_id": 1,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "ep_add_linear_projection": true,
+  "global_memory_size": 1,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_blocks": 200,
+  "max_position_embeddings": 1805,
+  "max_relative_position_embeddings": 8,
+  "model_type": "longelm",
+  "node_size": 1,
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 3,
+  "position_embedding_type": "mixed",
+  "torch_dtype": "float32",
+  "transformers_version": "4.34.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 50265
+}

checkpoint-18000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c21612b11c0bf3078ef59f425e373fa71ae79d928a1cc9d6ffa4b51506fccea
+size 1010431994

checkpoint-18000/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b1688c1c91c91fb3203af3bafd84865d15f17602048982f2b3b9c3cbc423e03
+size 505200682

checkpoint-18000/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8bd90b5e204dd40a8fc2417256c697b405e7cae6e0e41753eeff499f14b24a1
+size 15984

checkpoint-18000/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2c8a700bd1b3e41e8c0b90128fd817cc185d37f75db6af8a0672a7eb586430f
+size 15984

checkpoint-18000/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e2dc471de26d6827f15636beab38e8192e80c633cda0fd48b4de019c3c783bd
+size 15984

checkpoint-18000/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:206c61528118da9fd115f47182a0d29621365ea8799940d8756d3ebfb4c9b48f
+size 15984

checkpoint-18000/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eaa3e01685d79ec59eb3f3c7b71d847a081cb9c2a13e34f0c34a1e65a9c3ab31
+size 15984

checkpoint-18000/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17279ff1e099ae6ef523720a62084c9979d05b7702b3143563a83167868b44e8
+size 15984

checkpoint-18000/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9ad6be613d9bbdaf45e8850ed4a7896b4d1b7700f79e369893067c3ae458b89
+size 15984

checkpoint-18000/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:721e902be206a88094c65aff8adfa6d2673d92350e66dfa7b111a68ba8629142
+size 15984

checkpoint-18000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4047d89d88f030507a1a864e13de23ba5d1b052df50a4787936343b360d5dc71
+size 1064

checkpoint-18000/special_tokens_map.json

@@ -0,0 +1,37 @@
+{
+  "cls_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-18000/tokenizer_config.json

@@ -0,0 +1,60 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[PAD]",
+      "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
+    },
+    "30522": {
+      "content": "<INST>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "tokenizer_class": "LongelmTokenizer",
+  "unk_token": "[UNK]"
+}

checkpoint-18000/trainer_state.json

@@ -0,0 +1,307 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.3838835231368027,
+  "eval_steps": 500,
+  "global_step": 18000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "ep_loss": 4.4238,
+      "epoch": 0.04,
+      "learning_rate": 2.5e-05,
+      "loss": 11.1467,
+      "mlm_loss": 6.7229,
+      "step": 500
+    },
+    {
+      "ep_loss": 0.4305,
+      "epoch": 0.08,
+      "learning_rate": 5e-05,
+      "loss": 2.9841,
+      "mlm_loss": 2.5537,
+      "step": 1000
+    },
+    {
+      "ep_loss": 0.3724,
+      "epoch": 0.12,
+      "learning_rate": 7.5e-05,
+      "loss": 2.2371,
+      "mlm_loss": 1.8647,
+      "step": 1500
+    },
+    {
+      "ep_loss": 0.364,
+      "epoch": 0.15,
+      "learning_rate": 0.0001,
+      "loss": 1.8477,
+      "mlm_loss": 1.4837,
+      "step": 2000
+    },
+    {
+      "ep_loss": 0.3678,
+      "epoch": 0.19,
+      "learning_rate": 0.00012495,
+      "loss": 1.5215,
+      "mlm_loss": 1.1538,
+      "step": 2500
+    },
+    {
+      "ep_loss": 0.3617,
+      "epoch": 0.23,
+      "learning_rate": 0.00014995,
+      "loss": 1.4119,
+      "mlm_loss": 1.0501,
+      "step": 3000
+    },
+    {
+      "ep_loss": 0.3336,
+      "epoch": 0.27,
+      "learning_rate": 0.0001749,
+      "loss": 1.3027,
+      "mlm_loss": 0.9691,
+      "step": 3500
+    },
+    {
+      "ep_loss": 0.3348,
+      "epoch": 0.31,
+      "learning_rate": 0.0001999,
+      "loss": 1.2441,
+      "mlm_loss": 0.9093,
+      "step": 4000
+    },
+    {
+      "ep_loss": 0.3348,
+      "epoch": 0.35,
+      "learning_rate": 0.0002249,
+      "loss": 1.1942,
+      "mlm_loss": 0.8594,
+      "step": 4500
+    },
+    {
+      "ep_loss": 0.3331,
+      "epoch": 0.38,
+      "learning_rate": 0.0002499,
+      "loss": 1.1466,
+      "mlm_loss": 0.8135,
+      "step": 5000
+    },
+    {
+      "ep_loss": 0.3268,
+      "epoch": 0.42,
+      "learning_rate": 0.00027489999999999996,
+      "loss": 1.1067,
+      "mlm_loss": 0.7799,
+      "step": 5500
+    },
+    {
+      "ep_loss": 0.3378,
+      "epoch": 0.46,
+      "learning_rate": 0.00029985,
+      "loss": 1.1007,
+      "mlm_loss": 0.7629,
+      "step": 6000
+    },
+    {
+      "ep_loss": 0.3033,
+      "epoch": 0.5,
+      "learning_rate": 0.00032485,
+      "loss": 1.028,
+      "mlm_loss": 0.7246,
+      "step": 6500
+    },
+    {
+      "ep_loss": 0.2848,
+      "epoch": 0.54,
+      "learning_rate": 0.00034985,
+      "loss": 0.9833,
+      "mlm_loss": 0.6985,
+      "step": 7000
+    },
+    {
+      "ep_loss": 0.316,
+      "epoch": 0.58,
+      "learning_rate": 0.00037485000000000003,
+      "loss": 1.0235,
+      "mlm_loss": 0.7076,
+      "step": 7500
+    },
+    {
+      "ep_loss": 0.2885,
+      "epoch": 0.62,
+      "learning_rate": 0.00039975,
+      "loss": 0.9673,
+      "mlm_loss": 0.6789,
+      "step": 8000
+    },
+    {
+      "ep_loss": 0.274,
+      "epoch": 0.65,
+      "learning_rate": 0.00042475000000000005,
+      "loss": 0.9266,
+      "mlm_loss": 0.6525,
+      "step": 8500
+    },
+    {
+      "ep_loss": 0.3008,
+      "epoch": 0.69,
+      "learning_rate": 0.0004497,
+      "loss": 0.9646,
+      "mlm_loss": 0.6638,
+      "step": 9000
+    },
+    {
+      "ep_loss": 0.2843,
+      "epoch": 0.73,
+      "learning_rate": 0.00047470000000000005,
+      "loss": 0.9349,
+      "mlm_loss": 0.6506,
+      "step": 9500
+    },
+    {
+      "ep_loss": 0.2693,
+      "epoch": 0.77,
+      "learning_rate": 0.0004997,
+      "loss": 0.8901,
+      "mlm_loss": 0.6208,
+      "step": 10000
+    },
+    {
+      "ep_loss": 0.2655,
+      "epoch": 0.81,
+      "learning_rate": 0.0004995159140796489,
+      "loss": 0.8783,
+      "mlm_loss": 0.6128,
+      "step": 10500
+    },
+    {
+      "ep_loss": 0.2711,
+      "epoch": 0.85,
+      "learning_rate": 0.0004990269285042333,
+      "loss": 0.8794,
+      "mlm_loss": 0.6083,
+      "step": 11000
+    },
+    {
+      "ep_loss": 0.2579,
+      "epoch": 0.88,
+      "learning_rate": 0.000498536962997805,
+      "loss": 0.8489,
+      "mlm_loss": 0.591,
+      "step": 11500
+    },
+    {
+      "ep_loss": 0.2607,
+      "epoch": 0.92,
+      "learning_rate": 0.0004980469974913766,
+      "loss": 0.8504,
+      "mlm_loss": 0.5898,
+      "step": 12000
+    },
+    {
+      "ep_loss": 0.2585,
+      "epoch": 0.96,
+      "learning_rate": 0.0004975570319849483,
+      "loss": 0.8338,
+      "mlm_loss": 0.5753,
+      "step": 12500
+    },
+    {
+      "ep_loss": 0.2584,
+      "epoch": 1.0,
+      "learning_rate": 0.0004970670664785199,
+      "loss": 0.8287,
+      "mlm_loss": 0.5703,
+      "step": 13000
+    },
+    {
+      "ep_loss": 0.2511,
+      "epoch": 1.04,
+      "learning_rate": 0.0004965771009720916,
+      "loss": 0.8047,
+      "mlm_loss": 0.5536,
+      "step": 13500
+    },
+    {
+      "ep_loss": 0.2476,
+      "epoch": 1.08,
+      "learning_rate": 0.0004960871354656633,
+      "loss": 0.7952,
+      "mlm_loss": 0.5476,
+      "step": 14000
+    },
+    {
+      "ep_loss": 0.2482,
+      "epoch": 1.11,
+      "learning_rate": 0.0004955971699592349,
+      "loss": 0.7894,
+      "mlm_loss": 0.5412,
+      "step": 14500
+    },
+    {
+      "ep_loss": 0.2424,
+      "epoch": 1.15,
+      "learning_rate": 0.0004951072044528065,
+      "loss": 0.7744,
+      "mlm_loss": 0.5321,
+      "step": 15000
+    },
+    {
+      "ep_loss": 0.2462,
+      "epoch": 1.19,
+      "learning_rate": 0.0004946182188773911,
+      "loss": 0.7849,
+      "mlm_loss": 0.5386,
+      "step": 15500
+    },
+    {
+      "ep_loss": 0.235,
+      "epoch": 1.23,
+      "learning_rate": 0.0004941282533709627,
+      "loss": 0.7563,
+      "mlm_loss": 0.5213,
+      "step": 16000
+    },
+    {
+      "ep_loss": 0.2354,
+      "epoch": 1.27,
+      "learning_rate": 0.0004936382878645343,
+      "loss": 0.7513,
+      "mlm_loss": 0.5159,
+      "step": 16500
+    },
+    {
+      "ep_loss": 0.2336,
+      "epoch": 1.31,
+      "learning_rate": 0.0004931493022891189,
+      "loss": 0.7454,
+      "mlm_loss": 0.5118,
+      "step": 17000
+    },
+    {
+      "ep_loss": 0.233,
+      "epoch": 1.35,
+      "learning_rate": 0.0004926603167137033,
+      "loss": 0.7403,
+      "mlm_loss": 0.5073,
+      "step": 17500
+    },
+    {
+      "ep_loss": 0.2345,
+      "epoch": 1.38,
+      "learning_rate": 0.0004921713311382879,
+      "loss": 0.7399,
+      "mlm_loss": 0.5054,
+      "step": 18000
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 520240,
+  "num_train_epochs": 40,
+  "save_steps": 500,
+  "total_flos": 1.7186710895103836e+19,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-18000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b26e03be70ebe5d40a81a81be94421c0578a9d8742b065bc029501df48632957
+size 4664