Upload folder using huggingface_hub

training_checkpoints/checkpoint-6430/config.json

@@ -0,0 +1,63 @@
+{
+  "_name_or_path": "openaccess-ai-collective/tiny-mistral",
+  "architectures": [
+    "MistralForSequenceClassification"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "dropout_p": 0.1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 512,
+  "id2label": {
+    "0": "Issue",
+    "1": "Court Discourse",
+    "2": "Conclusion",
+    "3": "Precedent Analysis",
+    "4": "Section Analysis",
+    "5": "Argument by Petitioner",
+    "6": "Fact",
+    "7": "Argument by Respondent",
+    "8": "Ratio",
+    "9": "Appellant",
+    "10": "Respondent",
+    "11": "Argument by Appellant",
+    "12": "Petitioner",
+    "13": "Judge",
+    "14": "Argument by Defendant"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "label2id": {
+    "Appellant": 9,
+    "Argument by Appellant": 11,
+    "Argument by Defendant": 14,
+    "Argument by Petitioner": 5,
+    "Argument by Respondent": 7,
+    "Conclusion": 2,
+    "Court Discourse": 1,
+    "Fact": 6,
+    "Issue": 0,
+    "Judge": 13,
+    "Petitioner": 12,
+    "Precedent Analysis": 3,
+    "Ratio": 8,
+    "Respondent": 10,
+    "Section Analysis": 4
+  },
+  "max_position_embeddings": 32768,
+  "model_type": "mistral",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 8,
+  "num_key_value_heads": 4,
+  "pad_token_id": 32000,
+  "problem_type": "single_label_classification",
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 10000.0,
+  "sliding_window": 4096,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.39.3",
+  "use_cache": true,
+  "vocab_size": 32001
+}

training_checkpoints/checkpoint-6430/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:306a7966d8573d680e91a9902801e83027de8abaa4ce5f1e49c2c9f29e9802ee
+size 791226496

training_checkpoints/checkpoint-6430/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c2426f7059ff7880ab0372ee554a632875cb064eff383f71ec2f57cb4a478f2
+size 1582499834

training_checkpoints/checkpoint-6430/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de773a2cb0ba4b7f7d26731893c8e6a45c1e35ab31738065907b40eff7c937a4
+size 14244

training_checkpoints/checkpoint-6430/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:daa0ff26e40b7215f2997b9d5366ab0589d2924039f96c767879887fd5ed8a5f
+size 1064

training_checkpoints/checkpoint-6430/trainer_state.json

@@ -0,0 +1,325 @@
+{
+  "best_metric": 0.6236987905363891,
+  "best_model_checkpoint": "tiny-mistral/checkpoint-5144",
+  "epoch": 10.0,
+  "eval_steps": 500,
+  "global_step": 6430,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.78,
+      "grad_norm": 13.069233894348145,
+      "learning_rate": 4.7413167444271646e-05,
+      "loss": 1.4479,
+      "step": 500
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6498838109992254,
+      "eval_f1_macro": 0.45467871431664447,
+      "eval_f1_micro": 0.6498838109992254,
+      "eval_f1_weighted": 0.6214028764904677,
+      "eval_loss": 1.118202805519104,
+      "eval_macro_fpr": 0.039027492204575735,
+      "eval_macro_sensitivity": 0.4743680961365506,
+      "eval_macro_specificity": 0.9731278674075015,
+      "eval_precision": 0.6258370043663808,
+      "eval_precision_macro": 0.4712458606651466,
+      "eval_recall": 0.6498838109992254,
+      "eval_recall_macro": 0.4743680961365506,
+      "eval_runtime": 52.8829,
+      "eval_samples_per_second": 24.412,
+      "eval_steps_per_second": 3.063,
+      "eval_weighted_fpr": 0.03705525495982948,
+      "eval_weighted_sensitivity": 0.6498838109992254,
+      "eval_weighted_specificity": 0.9470342001132973,
+      "step": 643
+    },
+    {
+      "epoch": 1.56,
+      "grad_norm": 42.20966720581055,
+      "learning_rate": 4.482115085536548e-05,
+      "loss": 0.8133,
+      "step": 1000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6986831913245546,
+      "eval_f1_macro": 0.5473679856051097,
+      "eval_f1_micro": 0.6986831913245546,
+      "eval_f1_weighted": 0.6969580552726153,
+      "eval_loss": 1.0854355096817017,
+      "eval_macro_fpr": 0.030488809177152586,
+      "eval_macro_sensitivity": 0.552765390116517,
+      "eval_macro_specificity": 0.9773020581125585,
+      "eval_precision": 0.7196764053623382,
+      "eval_precision_macro": 0.5876945028566362,
+      "eval_recall": 0.6986831913245546,
+      "eval_recall_macro": 0.552765390116517,
+      "eval_runtime": 93.8123,
+      "eval_samples_per_second": 13.762,
+      "eval_steps_per_second": 1.727,
+      "eval_weighted_fpr": 0.029883997848966736,
+      "eval_weighted_sensitivity": 0.6986831913245546,
+      "eval_weighted_specificity": 0.9608476803638234,
+      "step": 1286
+    },
+    {
+      "epoch": 2.33,
+      "grad_norm": 26.4113712310791,
+      "learning_rate": 4.223950233281493e-05,
+      "loss": 0.5592,
+      "step": 1500
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6986831913245546,
+      "eval_f1_macro": 0.6012751387344262,
+      "eval_f1_micro": 0.6986831913245546,
+      "eval_f1_weighted": 0.6997681579317561,
+      "eval_loss": 1.6113793849945068,
+      "eval_macro_fpr": 0.030358926871943602,
+      "eval_macro_sensitivity": 0.5880627120151358,
+      "eval_macro_specificity": 0.9773067719065487,
+      "eval_precision": 0.7107021565435693,
+      "eval_precision_macro": 0.6368062038606529,
+      "eval_recall": 0.6986831913245546,
+      "eval_recall_macro": 0.5880627120151358,
+      "eval_runtime": 96.7771,
+      "eval_samples_per_second": 13.34,
+      "eval_steps_per_second": 1.674,
+      "eval_weighted_fpr": 0.029883997848966736,
+      "eval_weighted_sensitivity": 0.6986831913245546,
+      "eval_weighted_specificity": 0.9609183872736765,
+      "step": 1929
+    },
+    {
+      "epoch": 3.11,
+      "grad_norm": 0.03971128165721893,
+      "learning_rate": 3.964748574390876e-05,
+      "loss": 0.3916,
+      "step": 2000
+    },
+    {
+      "epoch": 3.89,
+      "grad_norm": 13.533605575561523,
+      "learning_rate": 3.705546915500259e-05,
+      "loss": 0.2375,
+      "step": 2500
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6955848179705655,
+      "eval_f1_macro": 0.5698961254173763,
+      "eval_f1_micro": 0.6955848179705655,
+      "eval_f1_weighted": 0.6923294650516646,
+      "eval_loss": 1.7778925895690918,
+      "eval_macro_fpr": 0.030968414851007965,
+      "eval_macro_sensitivity": 0.5666678921166624,
+      "eval_macro_specificity": 0.9768109367511323,
+      "eval_precision": 0.7000704251293273,
+      "eval_precision_macro": 0.584042243736088,
+      "eval_recall": 0.6955848179705655,
+      "eval_recall_macro": 0.5666678921166624,
+      "eval_runtime": 94.9154,
+      "eval_samples_per_second": 13.602,
+      "eval_steps_per_second": 1.707,
+      "eval_weighted_fpr": 0.030312379483224065,
+      "eval_weighted_sensitivity": 0.6955848179705655,
+      "eval_weighted_specificity": 0.956579233296421,
+      "step": 2572
+    },
+    {
+      "epoch": 4.67,
+      "grad_norm": 0.2196340560913086,
+      "learning_rate": 3.446345256609643e-05,
+      "loss": 0.1586,
+      "step": 3000
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6948102246320682,
+      "eval_f1_macro": 0.5694948674292386,
+      "eval_f1_micro": 0.6948102246320682,
+      "eval_f1_weighted": 0.691666971595594,
+      "eval_loss": 2.175213575363159,
+      "eval_macro_fpr": 0.03156565589564158,
+      "eval_macro_sensitivity": 0.5798857481261962,
+      "eval_macro_specificity": 0.9769938659292313,
+      "eval_precision": 0.7011163809357478,
+      "eval_precision_macro": 0.5797124163320468,
+      "eval_recall": 0.6948102246320682,
+      "eval_recall_macro": 0.5798857481261962,
+      "eval_runtime": 96.1916,
+      "eval_samples_per_second": 13.421,
+      "eval_steps_per_second": 1.684,
+      "eval_weighted_fpr": 0.03042001235330451,
+      "eval_weighted_sensitivity": 0.6948102246320682,
+      "eval_weighted_specificity": 0.9600977643064038,
+      "step": 3215
+    },
+    {
+      "epoch": 5.44,
+      "grad_norm": 0.0076691824942827225,
+      "learning_rate": 3.1881804043545884e-05,
+      "loss": 0.0956,
+      "step": 3500
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.7079783113865221,
+      "eval_f1_macro": 0.6114593419504739,
+      "eval_f1_micro": 0.7079783113865221,
+      "eval_f1_weighted": 0.710498142821041,
+      "eval_loss": 2.326085329055786,
+      "eval_macro_fpr": 0.029146593037922882,
+      "eval_macro_sensitivity": 0.6191389353797161,
+      "eval_macro_specificity": 0.9781721436893602,
+      "eval_precision": 0.7213132077015695,
+      "eval_precision_macro": 0.6169445235048693,
+      "eval_recall": 0.7079783113865221,
+      "eval_recall_macro": 0.6191389353797161,
+      "eval_runtime": 108.9589,
+      "eval_samples_per_second": 11.849,
+      "eval_steps_per_second": 1.487,
+      "eval_weighted_fpr": 0.028619145221285964,
+      "eval_weighted_sensitivity": 0.7079783113865221,
+      "eval_weighted_specificity": 0.9646038439538819,
+      "step": 3858
+    },
+    {
+      "epoch": 6.22,
+      "grad_norm": 0.003827617969363928,
+      "learning_rate": 2.928978745463971e-05,
+      "loss": 0.0623,
+      "step": 4000
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 0.007828492671251297,
+      "learning_rate": 2.669777086573354e-05,
+      "loss": 0.044,
+      "step": 4500
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.715724244771495,
+      "eval_f1_macro": 0.6014076418730009,
+      "eval_f1_micro": 0.715724244771495,
+      "eval_f1_weighted": 0.7131079309029965,
+      "eval_loss": 2.330761671066284,
+      "eval_macro_fpr": 0.028468870564731498,
+      "eval_macro_sensitivity": 0.5938828421663026,
+      "eval_macro_specificity": 0.9784561846662306,
+      "eval_precision": 0.7142878695652373,
+      "eval_precision_macro": 0.6184199211909682,
+      "eval_recall": 0.715724244771495,
+      "eval_recall_macro": 0.5938828421663026,
+      "eval_runtime": 106.6283,
+      "eval_samples_per_second": 12.107,
+      "eval_steps_per_second": 1.519,
+      "eval_weighted_fpr": 0.027587762158911525,
+      "eval_weighted_sensitivity": 0.715724244771495,
+      "eval_weighted_specificity": 0.9611185252219651,
+      "step": 4501
+    },
+    {
+      "epoch": 7.78,
+      "grad_norm": 0.0054510668851435184,
+      "learning_rate": 2.4105754276827372e-05,
+      "loss": 0.0212,
+      "step": 5000
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.7126258714175058,
+      "eval_f1_macro": 0.6236987905363891,
+      "eval_f1_micro": 0.7126258714175058,
+      "eval_f1_weighted": 0.7047150280912466,
+      "eval_loss": 2.5606603622436523,
+      "eval_macro_fpr": 0.029426308648760717,
+      "eval_macro_sensitivity": 0.6174588332474265,
+      "eval_macro_specificity": 0.9780489131013995,
+      "eval_precision": 0.7033345220138081,
+      "eval_precision_macro": 0.6494012124714204,
+      "eval_recall": 0.7126258714175058,
+      "eval_recall_macro": 0.6174588332474265,
+      "eval_runtime": 94.9282,
+      "eval_samples_per_second": 13.6,
+      "eval_steps_per_second": 1.707,
+      "eval_weighted_fpr": 0.027997886951928157,
+      "eval_weighted_sensitivity": 0.7126258714175058,
+      "eval_weighted_specificity": 0.9581078251034848,
+      "step": 5144
+    },
+    {
+      "epoch": 8.55,
+      "grad_norm": 0.002751357154920697,
+      "learning_rate": 2.1513737687921205e-05,
+      "loss": 0.0183,
+      "step": 5500
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.7118512780790085,
+      "eval_f1_macro": 0.5935171263710702,
+      "eval_f1_micro": 0.7118512780790084,
+      "eval_f1_weighted": 0.7088386468311138,
+      "eval_loss": 2.640535354614258,
+      "eval_macro_fpr": 0.02906658876848228,
+      "eval_macro_sensitivity": 0.5850361401718444,
+      "eval_macro_specificity": 0.978116781696577,
+      "eval_precision": 0.7091923605282856,
+      "eval_precision_macro": 0.6132653330348934,
+      "eval_recall": 0.7118512780790085,
+      "eval_recall_macro": 0.5850361401718444,
+      "eval_runtime": 97.5564,
+      "eval_samples_per_second": 13.233,
+      "eval_steps_per_second": 1.661,
+      "eval_weighted_fpr": 0.02810092158936395,
+      "eval_weighted_sensitivity": 0.7118512780790085,
+      "eval_weighted_specificity": 0.9599004473696481,
+      "step": 5787
+    },
+    {
+      "epoch": 9.33,
+      "grad_norm": 0.002242760267108679,
+      "learning_rate": 1.8921721099015034e-05,
+      "loss": 0.0145,
+      "step": 6000
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.7087529047250194,
+      "eval_f1_macro": 0.6038805077899971,
+      "eval_f1_micro": 0.7087529047250194,
+      "eval_f1_weighted": 0.7051494126143734,
+      "eval_loss": 2.7267849445343018,
+      "eval_macro_fpr": 0.0296548766519362,
+      "eval_macro_sensitivity": 0.594490599401291,
+      "eval_macro_specificity": 0.9777462293254275,
+      "eval_precision": 0.7057517403420687,
+      "eval_precision_macro": 0.6235388470549638,
+      "eval_recall": 0.7087529047250194,
+      "eval_recall_macro": 0.594490599401291,
+      "eval_runtime": 109.4092,
+      "eval_samples_per_second": 11.8,
+      "eval_steps_per_second": 1.481,
+      "eval_weighted_fpr": 0.028515091763992112,
+      "eval_weighted_sensitivity": 0.7087529047250194,
+      "eval_weighted_specificity": 0.9574405351563919,
+      "step": 6430
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 9645,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 15,
+  "save_steps": 500,
+  "total_flos": 2.863522174205952e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

training_checkpoints/checkpoint-6430/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae13135576265f1180bcf8daa3dbfb279a0746197d617e38954842ee54fec7e1
+size 4920