Adding model checkpoints and config files

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-0/config.json

@@ -0,0 +1,38 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "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,
+  "reorder_and_upcast_attn": true,
+  "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": true,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.2,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 1024
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.18.0",
+  "use_cache": false,
+  "vocab_size": 50257
+}

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-0/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b4794ea0b9a495e87fc34e316efe1b585527d6e3d15e4d093362ab24ee74ab3f
+size 510396521

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-0/special_tokens_map.json

@@ -0,0 +1 @@
+{}

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-0/tokenizer_config.json

@@ -0,0 +1 @@
+{"tokenizer_class": "PassthroughTokenizer"}

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-0/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfe85938b028b043a65c4e8b12b6788c64a3e8013a7acfbf6b8fb24e8dd9b106
+size 3183

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-100/config.json

@@ -0,0 +1,38 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "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,
+  "reorder_and_upcast_attn": true,
+  "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": true,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.2,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 1024
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.18.0",
+  "use_cache": false,
+  "vocab_size": 50257
+}

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-100/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3892e883c516bec96e17c271ec326bd599d52320b9876655707e3b9f15e0e446
+size 995603825

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-100/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e94ea2560c106c3d3ce3645948e160284e7622d1b2958f0ea3fccac2fd90e5e5
+size 510396521

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-100/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df6fc9ae57df51ecae22fa6bfacbb0bfa78885dfa78bf2f33e5ce6822acb6264
+size 14567

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-100/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13a3423b2fe42f204bc8fe2c666ff379f9fd753a0f13613064a5e71e86b519e8
+size 559

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-100/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75e9b9d31d11c624d89b0c04ad496adf4b5addd3e703848d2583972c703e8da6
+size 623

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-100/special_tokens_map.json

@@ -0,0 +1 @@
+{}

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-100/tokenizer_config.json

@@ -0,0 +1 @@
+{"tokenizer_class": "PassthroughTokenizer"}

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-100/trainer_state.json

@@ -0,0 +1,34 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.015833333333333,
+  "global_step": 100,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9999999999999996e-06,
+      "loss": 11.0217,
+      "step": 1
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.00025,
+      "loss": 8.0012,
+      "step": 50
+    },
+    {
+      "epoch": 3.02,
+      "learning_rate": 0.0005,
+      "loss": 5.7133,
+      "step": 100
+    }
+  ],
+  "max_steps": 1200,
+  "num_train_epochs": 9223372036854775807,
+  "total_flos": 2.7526070403072e+16,
+  "trial_name": null,
+  "trial_params": null
+}

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-100/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfe85938b028b043a65c4e8b12b6788c64a3e8013a7acfbf6b8fb24e8dd9b106
+size 3183

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1000/config.json

@@ -0,0 +1,38 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "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,
+  "reorder_and_upcast_attn": true,
+  "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": true,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.2,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 1024
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.18.0",
+  "use_cache": false,
+  "vocab_size": 50257
+}

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ff2b79960f69b7a050c0254caa0544ac4f5fa99793b756bc83414da5b894f7a
+size 995604017

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1000/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d1ebb8aa1bad585e70c3f9299975ca07c45f9ca2ab67a0b2142dbb3c6a16b995
+size 510396521

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b659b55d858751773a2297a8af7a96675a48b76cb05008e3e070fb8be589cdc8
+size 14567

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1000/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f810fc7b695697c440d8985f6042b4ba23a9e1027604c265718b518ca29f1b2b
+size 559

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0691206f4bd9ca409d6e7104087a4e0eb05df8f8f555a400f6ecc532edba52d8
+size 623

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1000/special_tokens_map.json

@@ -0,0 +1 @@
+{}

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1000/tokenizer_config.json

@@ -0,0 +1 @@
+{"tokenizer_class": "PassthroughTokenizer"}

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1000/trainer_state.json

@@ -0,0 +1,158 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 37.00083333333333,
+  "global_step": 1000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9999999999999996e-06,
+      "loss": 11.0217,
+      "step": 1
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.00025,
+      "loss": 8.0012,
+      "step": 50
+    },
+    {
+      "epoch": 3.02,
+      "learning_rate": 0.0005,
+      "loss": 5.7133,
+      "step": 100
+    },
+    {
+      "epoch": 5.01,
+      "learning_rate": 0.0005833333333333333,
+      "loss": 5.1467,
+      "step": 150
+    },
+    {
+      "epoch": 7.01,
+      "learning_rate": 0.0005555555555555556,
+      "loss": 4.9057,
+      "step": 200
+    },
+    {
+      "epoch": 9.01,
+      "learning_rate": 0.0005277777777777777,
+      "loss": 4.7624,
+      "step": 250
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 0.0005,
+      "loss": 4.5953,
+      "step": 300
+    },
+    {
+      "epoch": 12.02,
+      "learning_rate": 0.00047222222222222224,
+      "loss": 4.3139,
+      "step": 350
+    },
+    {
+      "epoch": 14.02,
+      "learning_rate": 0.00044444444444444436,
+      "loss": 4.2007,
+      "step": 400
+    },
+    {
+      "epoch": 16.02,
+      "learning_rate": 0.00041666666666666664,
+      "loss": 4.0441,
+      "step": 450
+    },
+    {
+      "epoch": 18.01,
+      "learning_rate": 0.00038888888888888887,
+      "loss": 3.9169,
+      "step": 500
+    },
+    {
+      "epoch": 20.01,
+      "learning_rate": 0.0003611111111111111,
+      "loss": 3.8046,
+      "step": 550
+    },
+    {
+      "epoch": 22.0,
+      "learning_rate": 0.0003333333333333333,
+      "loss": 3.706,
+      "step": 600
+    },
+    {
+      "epoch": 24.0,
+      "learning_rate": 0.00030555555555555555,
+      "loss": 3.6192,
+      "step": 650
+    },
+    {
+      "epoch": 25.02,
+      "learning_rate": 0.0002777777777777778,
+      "loss": 3.4774,
+      "step": 700
+    },
+    {
+      "epoch": 27.02,
+      "learning_rate": 0.00025,
+      "loss": 3.4742,
+      "step": 750
+    },
+    {
+      "epoch": 29.01,
+      "learning_rate": 0.00022222222222222218,
+      "loss": 3.411,
+      "step": 800
+    },
+    {
+      "epoch": 31.01,
+      "learning_rate": 0.00019444444444444443,
+      "loss": 3.3501,
+      "step": 850
+    },
+    {
+      "epoch": 33.01,
+      "learning_rate": 0.00016666666666666666,
+      "loss": 3.2943,
+      "step": 900
+    },
+    {
+      "epoch": 35.0,
+      "learning_rate": 0.0001388888888888889,
+      "loss": 3.2435,
+      "step": 950
+    },
+    {
+      "epoch": 37.0,
+      "learning_rate": 0.00011111111111111109,
+      "loss": 3.1955,
+      "step": 1000
+    },
+    {
+      "epoch": 37.0,
+      "eval_loss": 4.1568193435668945,
+      "eval_runtime": 3.9982,
+      "eval_samples_per_second": 50.773,
+      "eval_steps_per_second": 3.251,
+      "step": 1000
+    },
+    {
+      "epoch": 37.0,
+      "eval_/scratch/xiulyang/multilingual-LM/training/multilingual_dataset.py_loss": 4.1568193435668945,
+      "eval_/scratch/xiulyang/multilingual-LM/training/multilingual_dataset.py_ppl": 63.86805684825764,
+      "eval_/scratch/xiulyang/multilingual-LM/training/multilingual_dataset.py_runtime": 3.9982,
+      "eval_/scratch/xiulyang/multilingual-LM/training/multilingual_dataset.py_samples_per_second": 50.773,
+      "step": 1000
+    }
+  ],
+  "max_steps": 1200,
+  "num_train_epochs": 9223372036854775807,
+  "total_flos": 2.77056825458688e+17,
+  "trial_name": null,
+  "trial_params": null
+}

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfe85938b028b043a65c4e8b12b6788c64a3e8013a7acfbf6b8fb24e8dd9b106
+size 3183

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1100/config.json

@@ -0,0 +1,38 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "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,
+  "reorder_and_upcast_attn": true,
+  "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": true,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.2,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 1024
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.18.0",
+  "use_cache": false,
+  "vocab_size": 50257
+}

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1100/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5312dade7d066e5cd557a5df7c7de05c16bb06ea333b77ac35aeaeaf5693750a
+size 995604017

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1100/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a31698ef1c41ffaf01df801a9e92d3cfdeea227b3e19b8bf0f9ef85c4045b5a7
+size 510396521

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1100/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3cbf51f998b1b39c0b0c0dfb3ea6916e4c4bf96f57fe235ab93b204ad3bcc5d
+size 14567

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1100/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9fb16c30b686aa43e110b0d33f9d46bf3127b7124542ca8dc34831233d4675a0
+size 559

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1100/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f1b7713e4bb40428f29080b7d08d4a52f779ac863737861e4724292b2cf6c59
+size 623

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1100/special_tokens_map.json

@@ -0,0 +1 @@
+{}

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1100/tokenizer_config.json

@@ -0,0 +1 @@
+{"tokenizer_class": "PassthroughTokenizer"}

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1100/trainer_state.json

@@ -0,0 +1,170 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 40.016666666666666,
+  "global_step": 1100,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9999999999999996e-06,
+      "loss": 11.0217,
+      "step": 1
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.00025,
+      "loss": 8.0012,
+      "step": 50
+    },
+    {
+      "epoch": 3.02,
+      "learning_rate": 0.0005,
+      "loss": 5.7133,
+      "step": 100
+    },
+    {
+      "epoch": 5.01,
+      "learning_rate": 0.0005833333333333333,
+      "loss": 5.1467,
+      "step": 150
+    },
+    {
+      "epoch": 7.01,
+      "learning_rate": 0.0005555555555555556,
+      "loss": 4.9057,
+      "step": 200
+    },
+    {
+      "epoch": 9.01,
+      "learning_rate": 0.0005277777777777777,
+      "loss": 4.7624,
+      "step": 250
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 0.0005,
+      "loss": 4.5953,
+      "step": 300
+    },
+    {
+      "epoch": 12.02,
+      "learning_rate": 0.00047222222222222224,
+      "loss": 4.3139,
+      "step": 350
+    },
+    {
+      "epoch": 14.02,
+      "learning_rate": 0.00044444444444444436,
+      "loss": 4.2007,
+      "step": 400
+    },
+    {
+      "epoch": 16.02,
+      "learning_rate": 0.00041666666666666664,
+      "loss": 4.0441,
+      "step": 450
+    },
+    {
+      "epoch": 18.01,
+      "learning_rate": 0.00038888888888888887,
+      "loss": 3.9169,
+      "step": 500
+    },
+    {
+      "epoch": 20.01,
+      "learning_rate": 0.0003611111111111111,
+      "loss": 3.8046,
+      "step": 550
+    },
+    {
+      "epoch": 22.0,
+      "learning_rate": 0.0003333333333333333,
+      "loss": 3.706,
+      "step": 600
+    },
+    {
+      "epoch": 24.0,
+      "learning_rate": 0.00030555555555555555,
+      "loss": 3.6192,
+      "step": 650
+    },
+    {
+      "epoch": 25.02,
+      "learning_rate": 0.0002777777777777778,
+      "loss": 3.4774,
+      "step": 700
+    },
+    {
+      "epoch": 27.02,
+      "learning_rate": 0.00025,
+      "loss": 3.4742,
+      "step": 750
+    },
+    {
+      "epoch": 29.01,
+      "learning_rate": 0.00022222222222222218,
+      "loss": 3.411,
+      "step": 800
+    },
+    {
+      "epoch": 31.01,
+      "learning_rate": 0.00019444444444444443,
+      "loss": 3.3501,
+      "step": 850
+    },
+    {
+      "epoch": 33.01,
+      "learning_rate": 0.00016666666666666666,
+      "loss": 3.2943,
+      "step": 900
+    },
+    {
+      "epoch": 35.0,
+      "learning_rate": 0.0001388888888888889,
+      "loss": 3.2435,
+      "step": 950
+    },
+    {
+      "epoch": 37.0,
+      "learning_rate": 0.00011111111111111109,
+      "loss": 3.1955,
+      "step": 1000
+    },
+    {
+      "epoch": 37.0,
+      "eval_loss": 4.1568193435668945,
+      "eval_runtime": 3.9982,
+      "eval_samples_per_second": 50.773,
+      "eval_steps_per_second": 3.251,
+      "step": 1000
+    },
+    {
+      "epoch": 37.0,
+      "eval_/scratch/xiulyang/multilingual-LM/training/multilingual_dataset.py_loss": 4.1568193435668945,
+      "eval_/scratch/xiulyang/multilingual-LM/training/multilingual_dataset.py_ppl": 63.86805684825764,
+      "eval_/scratch/xiulyang/multilingual-LM/training/multilingual_dataset.py_runtime": 3.9982,
+      "eval_/scratch/xiulyang/multilingual-LM/training/multilingual_dataset.py_samples_per_second": 50.773,
+      "step": 1000
+    },
+    {
+      "epoch": 38.02,
+      "learning_rate": 8.333333333333333e-05,
+      "loss": 3.0973,
+      "step": 1050
+    },
+    {
+      "epoch": 40.02,
+      "learning_rate": 5.5555555555555545e-05,
+      "loss": 3.1214,
+      "step": 1100
+    }
+  ],
+  "max_steps": 1200,
+  "num_train_epochs": 9223372036854775807,
+  "total_flos": 3.0458289586176e+17,
+  "trial_name": null,
+  "trial_params": null
+}

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1100/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfe85938b028b043a65c4e8b12b6788c64a3e8013a7acfbf6b8fb24e8dd9b106
+size 3183

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1200/config.json

@@ -0,0 +1,38 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "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,
+  "reorder_and_upcast_attn": true,
+  "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": true,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.2,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 1024
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.18.0",
+  "use_cache": false,
+  "vocab_size": 50257
+}

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1200/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01d87af56b6da9fc268635e1dfe32d11b8bd98d82bd6c4270bbc7bb5c3a690d4
+size 995604017

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1200/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b6b16cad85cae699e713528f709cb28175652190ccf93624f41b01a74ac113e
+size 510396521

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1200/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:22724403665e347fa884491881d38603434d9ece1ef30562ba13d8e7ca7c236d
+size 14567

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1200/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19c7277eaca0850ae3e9b6790b3d002d820169cce0671185e672c28c8ae8e056
+size 559

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1200/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:935a8fb09a6e9698d9894853b05e181b3f56098deaaecddde08e55f06bf000c4
+size 623

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1200/special_tokens_map.json

@@ -0,0 +1 @@
+{}

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1200/tokenizer_config.json

@@ -0,0 +1 @@
+{"tokenizer_class": "PassthroughTokenizer"}

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1200/trainer_state.json

@@ -0,0 +1,182 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 44.01,
+  "global_step": 1200,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9999999999999996e-06,
+      "loss": 11.0217,
+      "step": 1
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.00025,
+      "loss": 8.0012,
+      "step": 50
+    },
+    {
+      "epoch": 3.02,
+      "learning_rate": 0.0005,
+      "loss": 5.7133,
+      "step": 100
+    },
+    {
+      "epoch": 5.01,
+      "learning_rate": 0.0005833333333333333,
+      "loss": 5.1467,
+      "step": 150
+    },
+    {
+      "epoch": 7.01,
+      "learning_rate": 0.0005555555555555556,
+      "loss": 4.9057,
+      "step": 200
+    },
+    {
+      "epoch": 9.01,
+      "learning_rate": 0.0005277777777777777,
+      "loss": 4.7624,
+      "step": 250
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 0.0005,
+      "loss": 4.5953,
+      "step": 300
+    },
+    {
+      "epoch": 12.02,
+      "learning_rate": 0.00047222222222222224,
+      "loss": 4.3139,
+      "step": 350
+    },
+    {
+      "epoch": 14.02,
+      "learning_rate": 0.00044444444444444436,
+      "loss": 4.2007,
+      "step": 400
+    },
+    {
+      "epoch": 16.02,
+      "learning_rate": 0.00041666666666666664,
+      "loss": 4.0441,
+      "step": 450
+    },
+    {
+      "epoch": 18.01,
+      "learning_rate": 0.00038888888888888887,
+      "loss": 3.9169,
+      "step": 500
+    },
+    {
+      "epoch": 20.01,
+      "learning_rate": 0.0003611111111111111,
+      "loss": 3.8046,
+      "step": 550
+    },
+    {
+      "epoch": 22.0,
+      "learning_rate": 0.0003333333333333333,
+      "loss": 3.706,
+      "step": 600
+    },
+    {
+      "epoch": 24.0,
+      "learning_rate": 0.00030555555555555555,
+      "loss": 3.6192,
+      "step": 650
+    },
+    {
+      "epoch": 25.02,
+      "learning_rate": 0.0002777777777777778,
+      "loss": 3.4774,
+      "step": 700
+    },
+    {
+      "epoch": 27.02,
+      "learning_rate": 0.00025,
+      "loss": 3.4742,
+      "step": 750
+    },
+    {
+      "epoch": 29.01,
+      "learning_rate": 0.00022222222222222218,
+      "loss": 3.411,
+      "step": 800
+    },
+    {
+      "epoch": 31.01,
+      "learning_rate": 0.00019444444444444443,
+      "loss": 3.3501,
+      "step": 850
+    },
+    {
+      "epoch": 33.01,
+      "learning_rate": 0.00016666666666666666,
+      "loss": 3.2943,
+      "step": 900
+    },
+    {
+      "epoch": 35.0,
+      "learning_rate": 0.0001388888888888889,
+      "loss": 3.2435,
+      "step": 950
+    },
+    {
+      "epoch": 37.0,
+      "learning_rate": 0.00011111111111111109,
+      "loss": 3.1955,
+      "step": 1000
+    },
+    {
+      "epoch": 37.0,
+      "eval_loss": 4.1568193435668945,
+      "eval_runtime": 3.9982,
+      "eval_samples_per_second": 50.773,
+      "eval_steps_per_second": 3.251,
+      "step": 1000
+    },
+    {
+      "epoch": 37.0,
+      "eval_/scratch/xiulyang/multilingual-LM/training/multilingual_dataset.py_loss": 4.1568193435668945,
+      "eval_/scratch/xiulyang/multilingual-LM/training/multilingual_dataset.py_ppl": 63.86805684825764,
+      "eval_/scratch/xiulyang/multilingual-LM/training/multilingual_dataset.py_runtime": 3.9982,
+      "eval_/scratch/xiulyang/multilingual-LM/training/multilingual_dataset.py_samples_per_second": 50.773,
+      "step": 1000
+    },
+    {
+      "epoch": 38.02,
+      "learning_rate": 8.333333333333333e-05,
+      "loss": 3.0973,
+      "step": 1050
+    },
+    {
+      "epoch": 40.02,
+      "learning_rate": 5.5555555555555545e-05,
+      "loss": 3.1214,
+      "step": 1100
+    },
+    {
+      "epoch": 42.01,
+      "learning_rate": 2.7777777777777772e-05,
+      "loss": 3.0907,
+      "step": 1150
+    },
+    {
+      "epoch": 44.01,
+      "learning_rate": 0.0,
+      "loss": 3.0653,
+      "step": 1200
+    }
+  ],
+  "max_steps": 1200,
+  "num_train_epochs": 9223372036854775807,
+  "total_flos": 3.32365555040256e+17,
+  "trial_name": null,
+  "trial_params": null
+}

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-1200/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfe85938b028b043a65c4e8b12b6788c64a3e8013a7acfbf6b8fb24e8dd9b106
+size 3183

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-200/config.json

@@ -0,0 +1,38 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "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,
+  "reorder_and_upcast_attn": true,
+  "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": true,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.2,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 1024
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.18.0",
+  "use_cache": false,
+  "vocab_size": 50257
+}

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-200/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32e5f22abe3cb0150546e63e1f90e1ca7e307db1d4233880c96ce0dc4b4bac7f
+size 995603825

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-200/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66d408c99d7d507fffcdf2d4cbcb652b92bd635273775d23e927998100b00f54
+size 510396521

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-200/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a45350f255628f9e40f118bea3dc7b01319b280bcb1f46c7b5d3c3a962da6d42
+size 14567

multilingual_models/perturb_det_adj_np_num_en_EN_randinit/babylm_perturb_det_adj_np_num_en_EN_randinit_seed53/runs/perturb_det_adj_np_num_en_EN_randinit_seed53/checkpoint-200/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb6982c29cd162f49aeb531674acf574eccd46a8f556bec596040d7c3b95200a
+size 559