test

data/csv/default-ebd4ee978eb95743/0.0.0/6b34fb8fcf56f7c8ba51dc895bfa2bfbe43546f190a60fcf74bb5e8afdcc2317/cache-c7cc71d1ae90354b.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1b8096b9de9b0e608bec3162bdddc28dfdb1e248225ac935fdaa12e0c23fba4
+size 38479112

result/sup-simcse-pere-roberta-scand-base-gpu/config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "result/sup-simcse-pere-roberta-scand-base-gpu",
+  "architectures": [
+    "RobertaForCL"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.2.1",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 250002
+}

result/sup-simcse-pere-roberta-scand-base-gpu/eval_results.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b95ac1ddc5795bf753904df5cc4baefe78270c1e02d02e4265aefc57c20d0d3
+size 286

result/sup-simcse-pere-roberta-scand-base-gpu/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fedf5b216252cb71ee0aaef3af4adcace5718203a91acb04f7962d90b4aa793a
+size 2224474093

result/sup-simcse-pere-roberta-scand-base-gpu/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae43dc9332417216710145e094741ab5af4deccabf16db251fcb893d3b13d002
+size 1112264504

result/sup-simcse-pere-roberta-scand-base-gpu/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ff735d7e71cf691b44d5a526cf5e2c54968b571005f1b9c5f542094a4195bbb
+size 623

result/sup-simcse-pere-roberta-scand-base-gpu/sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
+size 5069051

result/sup-simcse-pere-roberta-scand-base-gpu/special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": "<mask>"}

result/sup-simcse-pere-roberta-scand-base-gpu/tokenizer_config.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": "<mask>", "special_tokens_map_file": null, "name_or_path": "pere/roberta-base-exp"}

result/sup-simcse-pere-roberta-scand-base-gpu/train_results.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e28e39647a91e03d3079926faf9aee9a3cfb1314259b1715513eef933be9596
+size 71

result/sup-simcse-pere-roberta-scand-base-gpu/trainer_state.json

@@ -0,0 +1,124 @@
+{
+  "best_metric": 0.7855053623703318,
+  "best_model_checkpoint": "result/sup-simcse-pere-roberta-scand-base-gpu",
+  "epoch": 3.0,
+  "global_step": 1503,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.25,
+      "eval_avg_sts": 0.7174999106521781,
+      "eval_sickr_spearman": 0.6692206504832229,
+      "eval_stsb_spearman": 0.7657791708211333,
+      "step": 125
+    },
+    {
+      "epoch": 0.5,
+      "eval_avg_sts": 0.7052877421829599,
+      "eval_sickr_spearman": 0.6512401596287639,
+      "eval_stsb_spearman": 0.7593353247371558,
+      "step": 250
+    },
+    {
+      "epoch": 0.75,
+      "eval_avg_sts": 0.7324103077852453,
+      "eval_sickr_spearman": 0.6825378497301569,
+      "eval_stsb_spearman": 0.7822827658403337,
+      "step": 375
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.33666001330672e-05,
+      "loss": 0.8632,
+      "step": 500
+    },
+    {
+      "epoch": 1.0,
+      "eval_avg_sts": 0.7292681219582265,
+      "eval_sickr_spearman": 0.6828054309962686,
+      "eval_stsb_spearman": 0.7757308129201845,
+      "step": 500
+    },
+    {
+      "epoch": 1.25,
+      "eval_avg_sts": 0.7352730226575863,
+      "eval_sickr_spearman": 0.6886684434555147,
+      "eval_stsb_spearman": 0.781877601859658,
+      "step": 625
+    },
+    {
+      "epoch": 1.5,
+      "eval_avg_sts": 0.7386464094543315,
+      "eval_sickr_spearman": 0.6931255855756673,
+      "eval_stsb_spearman": 0.7841672333329957,
+      "step": 750
+    },
+    {
+      "epoch": 1.75,
+      "eval_avg_sts": 0.7378583784642221,
+      "eval_sickr_spearman": 0.6902113945581125,
+      "eval_stsb_spearman": 0.7855053623703318,
+      "step": 875
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.67332002661344e-05,
+      "loss": 0.53,
+      "step": 1000
+    },
+    {
+      "epoch": 2.0,
+      "eval_avg_sts": 0.7315130779911422,
+      "eval_sickr_spearman": 0.6861851394489614,
+      "eval_stsb_spearman": 0.7768410165333229,
+      "step": 1000
+    },
+    {
+      "epoch": 2.25,
+      "eval_avg_sts": 0.7303559585224864,
+      "eval_sickr_spearman": 0.6827572077704247,
+      "eval_stsb_spearman": 0.7779547092745481,
+      "step": 1125
+    },
+    {
+      "epoch": 2.5,
+      "eval_avg_sts": 0.7338041675107871,
+      "eval_sickr_spearman": 0.6878975923085344,
+      "eval_stsb_spearman": 0.77971074271304,
+      "step": 1250
+    },
+    {
+      "epoch": 2.74,
+      "eval_avg_sts": 0.7292608246481702,
+      "eval_sickr_spearman": 0.6831592280894618,
+      "eval_stsb_spearman": 0.7753624212068787,
+      "step": 1375
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 9.98003992015968e-08,
+      "loss": 0.4249,
+      "step": 1500
+    },
+    {
+      "epoch": 2.99,
+      "eval_avg_sts": 0.7335215968699524,
+      "eval_sickr_spearman": 0.6882147416713302,
+      "eval_stsb_spearman": 0.7788284520685745,
+      "step": 1500
+    },
+    {
+      "epoch": 3.0,
+      "step": 1503,
+      "train_runtime": 1369.7623,
+      "train_samples_per_second": 1.097
+    }
+  ],
+  "max_steps": 1503,
+  "num_train_epochs": 3,
+  "total_flos": 61539164830826496,
+  "trial_name": null,
+  "trial_params": null
+}

result/sup-simcse-pere-roberta-scand-base-gpu/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a939f3a66acb0551fdbeda2860150a40d9388ad85c425802878c59e648a947d
+size 2095