Upload 8 files

config.json

@@ -0,0 +1,43 @@
+{
+ "_name_or_path": "xlnet-base-cased",
+ "architectures": [
+ "XLNetLMHeadModel"
+ ],
+ "attn_type": "bi",
+ "bi_data": false,
+ "bos_token_id": 1,
+ "clamp_len": -1,
+ "d_head": 64,
+ "d_inner": 3072,
+ "d_model": 768,
+ "dropout": 0.1,
+ "end_n_top": 5,
+ "eos_token_id": 2,
+ "ff_activation": "gelu",
+ "initializer_range": 0.02,
+ "layer_norm_eps": 1e-12,
+ "mem_len": null,
+ "model_type": "xlnet",
+ "n_head": 12,
+ "n_layer": 12,
+ "pad_token_id": 5,
+ "reuse_len": null,
+ "same_length": false,
+ "start_n_top": 5,
+ "summary_activation": "tanh",
+ "summary_last_dropout": 0.1,
+ "summary_type": "last",
+ "summary_use_proj": true,
+ "task_specific_params": {
+ "text-generation": {
+ "do_sample": true,
+ "max_length": 250
+ }
+ },
+ "torch_dtype": "float32",
+ "transformers_version": "4.37.2",
+ "untie_r": true,
+ "use_mems_eval": true,
+ "use_mems_train": false,
+ "vocab_size": 32000
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+ "_from_model_config": true,
+ "bos_token_id": 1,
+ "eos_token_id": 2,
+ "pad_token_id": 5,
+ "transformers_version": "4.37.2"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f79ba011086762a6dcecedcabee4950be955b06e582cca213b7477ad43a43f5d
+size 467023808

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c312c0f5bb1e715a4ab42bd3b9a8544f07644cab6d15fdac7b67090cb4d74026
+size 934176698

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eec6375636ca4fed14e8a00efc534e0e68668357cdd84bb5f236659ce9aa510
+size 14244

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1eca3c6ddfef1190391ae3bd6a8168be026c9c4ecc8d3231c7c3526e6c3acc8
+size 1064

trainer_state.json

@@ -0,0 +1,161 @@
+{
+ "best_metric": 9.289751324104145e-05,
+ "best_model_checkpoint": "data/further_pretraining/xlnet-base-cased/checkpoint-4706",
+ "epoch": 4.0,
+ "eval_steps": 500,
+ "global_step": 9412,
+ "is_hyper_param_search": false,
+ "is_local_process_zero": true,
+ "is_world_process_zero": true,
+ "log_history": [
+ {
+ "epoch": 0.21,
+ "learning_rate": 2.9362515937101572e-05,
+ "loss": 0.018,
+ "step": 500
+ },
+ {
+ "epoch": 0.42,
+ "learning_rate": 2.8725031874203144e-05,
+ "loss": 0.0003,
+ "step": 1000
+ },
+ {
+ "epoch": 0.64,
+ "learning_rate": 2.808754781130472e-05,
+ "loss": 0.0005,
+ "step": 1500
+ },
+ {
+ "epoch": 0.85,
+ "learning_rate": 2.7450063748406293e-05,
+ "loss": 0.0001,
+ "step": 2000
+ },
+ {
+ "epoch": 1.0,
+ "eval_loss": 0.00010193797788815573,
+ "eval_runtime": 228.2334,
+ "eval_samples_per_second": 329.847,
+ "eval_steps_per_second": 10.31,
+ "step": 2353
+ },
+ {
+ "epoch": 1.06,
+ "learning_rate": 2.6812579685507864e-05,
+ "loss": 0.0004,
+ "step": 2500
+ },
+ {
+ "epoch": 1.27,
+ "learning_rate": 2.6175095622609436e-05,
+ "loss": 0.0003,
+ "step": 3000
+ },
+ {
+ "epoch": 1.49,
+ "learning_rate": 2.5537611559711007e-05,
+ "loss": 0.0007,
+ "step": 3500
+ },
+ {
+ "epoch": 1.7,
+ "learning_rate": 2.490012749681258e-05,
+ "loss": 0.0005,
+ "step": 4000
+ },
+ {
+ "epoch": 1.91,
+ "learning_rate": 2.4262643433914153e-05,
+ "loss": 0.0007,
+ "step": 4500
+ },
+ {
+ "epoch": 2.0,
+ "eval_loss": 9.289751324104145e-05,
+ "eval_runtime": 228.3262,
+ "eval_samples_per_second": 329.712,
+ "eval_steps_per_second": 10.305,
+ "step": 4706
+ },
+ {
+ "epoch": 2.12,
+ "learning_rate": 2.3625159371015725e-05,
+ "loss": 0.0022,
+ "step": 5000
+ },
+ {
+ "epoch": 2.34,
+ "learning_rate": 2.2987675308117296e-05,
+ "loss": 0.0002,
+ "step": 5500
+ },
+ {
+ "epoch": 2.55,
+ "learning_rate": 2.235019124521887e-05,
+ "loss": 0.0001,
+ "step": 6000
+ },
+ {
+ "epoch": 2.76,
+ "learning_rate": 2.1712707182320442e-05,
+ "loss": 0.0,
+ "step": 6500
+ },
+ {
+ "epoch": 2.97,
+ "learning_rate": 2.1075223119422017e-05,
+ "loss": 0.0004,
+ "step": 7000
+ },
+ {
+ "epoch": 3.0,
+ "eval_loss": 0.00031714412034489214,
+ "eval_runtime": 228.2639,
+ "eval_samples_per_second": 329.802,
+ "eval_steps_per_second": 10.308,
+ "step": 7059
+ },
+ {
+ "epoch": 3.19,
+ "learning_rate": 2.043773905652359e-05,
+ "loss": 0.0002,
+ "step": 7500
+ },
+ {
+ "epoch": 3.4,
+ "learning_rate": 1.980025499362516e-05,
+ "loss": 0.0002,
+ "step": 8000
+ },
+ {
+ "epoch": 3.61,
+ "learning_rate": 1.916277093072673e-05,
+ "loss": 0.0015,
+ "step": 8500
+ },
+ {
+ "epoch": 3.82,
+ "learning_rate": 1.8525286867828303e-05,
+ "loss": 0.0,
+ "step": 9000
+ },
+ {
+ "epoch": 4.0,
+ "eval_loss": 0.00011609662760747597,
+ "eval_runtime": 228.3349,
+ "eval_samples_per_second": 329.7,
+ "eval_steps_per_second": 10.305,
+ "step": 9412
+ }
+ ],
+ "logging_steps": 500,
+ "max_steps": 23530,
+ "num_input_tokens_seen": 0,
+ "num_train_epochs": 10,
+ "save_steps": 500,
+ "total_flos": 2.1316818010374144e+16,
+ "train_batch_size": 32,
+ "trial_name": null,
+ "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:059aa0c3a615de0216062fe9a2bd35fc361e9e49cc50dffa65fee860213c8c4d
+size 4728