Token Classification
Transformers
Safetensors
deberta-v2
named-entity-recognition
ner
multilingual
slavic
mdeberta-v3
iob2
Eval Results (legacy)
Instructions to use ivlcic/snerta-8l-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use ivlcic/snerta-8l-base with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="ivlcic/snerta-8l-base")# Load model directly from transformers import AutoTokenizer, AutoModelForTokenClassification tokenizer = AutoTokenizer.from_pretrained("ivlcic/snerta-8l-base") model = AutoModelForTokenClassification.from_pretrained("ivlcic/snerta-8l-base", device_map="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "best_global_step": 64215, | |
| "best_metric": 0.9238630978137686, | |
| "best_model_checkpoint": "/workspace/cannopy/result/train/token/ner/ner.mdeberta3.b16.lr2e-05/checkpoint-64215", | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 85620, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.3367141485214233, | |
| "learning_rate": 1.6660179057999222e-05, | |
| "loss": 0.39717576459756776, | |
| "step": 4281 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_acc": 0.9832746262039067, | |
| "eval_f1": 0.8776248509468665, | |
| "eval_label.LOC.f1": 0.93492293505467, | |
| "eval_label.LOC.p": 0.923727710529741, | |
| "eval_label.LOC.r": 0.9463928523803173, | |
| "eval_label.MISC.f1": 0.7478156071105755, | |
| "eval_label.MISC.p": 0.7107674684994273, | |
| "eval_label.MISC.r": 0.7889383343928799, | |
| "eval_label.ORG.f1": 0.8458744915746658, | |
| "eval_label.ORG.p": 0.8004123711340206, | |
| "eval_label.ORG.r": 0.8968119513322039, | |
| "eval_label.PER.f1": 0.9085019077100281, | |
| "eval_label.PER.p": 0.9105339105339105, | |
| "eval_label.PER.r": 0.906478954173251, | |
| "eval_loss": 0.056813765317201614, | |
| "eval_p": 0.8554767533490938, | |
| "eval_r": 0.9009502468982116, | |
| "eval_runtime": 32.4812, | |
| "eval_samples_per_second": 555.891, | |
| "eval_steps_per_second": 17.395, | |
| "step": 4281 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.3472847938537598, | |
| "learning_rate": 1.9149375015531425e-05, | |
| "loss": 0.09596302698104633, | |
| "step": 8562 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_acc": 0.9856530176011911, | |
| "eval_f1": 0.8988005842779847, | |
| "eval_label.LOC.f1": 0.9451052150045746, | |
| "eval_label.LOC.p": 0.9266948609509164, | |
| "eval_label.LOC.r": 0.9642619015868783, | |
| "eval_label.MISC.f1": 0.7947612732095491, | |
| "eval_label.MISC.p": 0.8305613305613305, | |
| "eval_label.MISC.r": 0.7619198982835347, | |
| "eval_label.ORG.f1": 0.8627658769207479, | |
| "eval_label.ORG.p": 0.8483179517713605, | |
| "eval_label.ORG.r": 0.8777144617280148, | |
| "eval_label.PER.f1": 0.9268951784952354, | |
| "eval_label.PER.p": 0.9177580622447543, | |
| "eval_label.PER.r": 0.9362160609107887, | |
| "eval_loss": 0.05167277529835701, | |
| "eval_p": 0.8913007997388608, | |
| "eval_r": 0.9064276525996929, | |
| "eval_runtime": 32.474, | |
| "eval_samples_per_second": 556.014, | |
| "eval_steps_per_second": 17.399, | |
| "step": 8562 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.4700154662132263, | |
| "learning_rate": 1.808553465371139e-05, | |
| "loss": 0.06998056341793682, | |
| "step": 12843 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_acc": 0.9857843901985095, | |
| "eval_f1": 0.9082807731434385, | |
| "eval_label.LOC.f1": 0.9483767961681746, | |
| "eval_label.LOC.p": 0.946236559139785, | |
| "eval_label.LOC.r": 0.9505267368982531, | |
| "eval_label.MISC.f1": 0.8065855829130822, | |
| "eval_label.MISC.p": 0.7561179087875417, | |
| "eval_label.MISC.r": 0.8642720915448188, | |
| "eval_label.ORG.f1": 0.8929755803436841, | |
| "eval_label.ORG.p": 0.8743911439114391, | |
| "eval_label.ORG.r": 0.9123671646388418, | |
| "eval_label.PER.f1": 0.9285511766373687, | |
| "eval_label.PER.p": 0.916468448299986, | |
| "eval_label.PER.r": 0.9409567590863381, | |
| "eval_loss": 0.04598889499902725, | |
| "eval_p": 0.8910223144784639, | |
| "eval_r": 0.9262210050209553, | |
| "eval_runtime": 33.4856, | |
| "eval_samples_per_second": 539.217, | |
| "eval_steps_per_second": 16.873, | |
| "step": 12843 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.12420473247766495, | |
| "learning_rate": 1.7021694291891355e-05, | |
| "loss": 0.05369556821748715, | |
| "step": 17124 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_acc": 0.9874183368776566, | |
| "eval_f1": 0.9158554693944893, | |
| "eval_label.LOC.f1": 0.9499669093315686, | |
| "eval_label.LOC.p": 0.9429772697411641, | |
| "eval_label.LOC.r": 0.9570609414588612, | |
| "eval_label.MISC.f1": 0.8372469635627531, | |
| "eval_label.MISC.p": 0.8534169692967977, | |
| "eval_label.MISC.r": 0.8216783216783217, | |
| "eval_label.ORG.f1": 0.8995784402288467, | |
| "eval_label.ORG.p": 0.8798409659843911, | |
| "eval_label.ORG.r": 0.9202217772986293, | |
| "eval_label.PER.f1": 0.9292334219412314, | |
| "eval_label.PER.p": 0.9294337453291176, | |
| "eval_label.PER.r": 0.9290331848872289, | |
| "eval_loss": 0.04518148675560951, | |
| "eval_p": 0.9104104309319775, | |
| "eval_r": 0.9213660317855513, | |
| "eval_runtime": 32.8809, | |
| "eval_samples_per_second": 549.133, | |
| "eval_steps_per_second": 17.183, | |
| "step": 17124 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.2671431601047516, | |
| "learning_rate": 1.595785393007132e-05, | |
| "loss": 0.04225894363927273, | |
| "step": 21405 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_acc": 0.9871884348323494, | |
| "eval_f1": 0.9171679967260078, | |
| "eval_label.LOC.f1": 0.9532500165311116, | |
| "eval_label.LOC.p": 0.9454354669464848, | |
| "eval_label.LOC.r": 0.961194825976797, | |
| "eval_label.MISC.f1": 0.8348736906962415, | |
| "eval_label.MISC.p": 0.8099222952779438, | |
| "eval_label.MISC.r": 0.8614113159567705, | |
| "eval_label.ORG.f1": 0.901406320883109, | |
| "eval_label.ORG.p": 0.8853408584583395, | |
| "eval_label.ORG.r": 0.9180656091175111, | |
| "eval_label.PER.f1": 0.9312139140352127, | |
| "eval_label.PER.p": 0.9241652518392756, | |
| "eval_label.PER.r": 0.9383709237178566, | |
| "eval_loss": 0.048478372395038605, | |
| "eval_p": 0.904727302087118, | |
| "eval_r": 0.9299555998174198, | |
| "eval_runtime": 32.4315, | |
| "eval_samples_per_second": 556.742, | |
| "eval_steps_per_second": 17.421, | |
| "step": 21405 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.6828418374061584, | |
| "learning_rate": 1.4894013568251286e-05, | |
| "loss": 0.033062972968996365, | |
| "step": 25686 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_acc": 0.9869092680630479, | |
| "eval_f1": 0.9160358586169406, | |
| "eval_label.LOC.f1": 0.9566717536991572, | |
| "eval_label.LOC.p": 0.9520602218700476, | |
| "eval_label.LOC.r": 0.9613281770902787, | |
| "eval_label.MISC.f1": 0.8213117307991898, | |
| "eval_label.MISC.p": 0.8053773296669722, | |
| "eval_label.MISC.r": 0.8378893833439288, | |
| "eval_label.ORG.f1": 0.9023171285779191, | |
| "eval_label.ORG.p": 0.8875316549977655, | |
| "eval_label.ORG.r": 0.9176035730787001, | |
| "eval_label.PER.f1": 0.9286171802718668, | |
| "eval_label.PER.p": 0.9201692524682652, | |
| "eval_label.PER.r": 0.9372216635540871, | |
| "eval_loss": 0.053501419723033905, | |
| "eval_p": 0.9058341447581953, | |
| "eval_r": 0.9264699780073862, | |
| "eval_runtime": 32.9114, | |
| "eval_samples_per_second": 548.625, | |
| "eval_steps_per_second": 17.167, | |
| "step": 25686 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.4229598343372345, | |
| "learning_rate": 1.3830173206431251e-05, | |
| "loss": 0.025886756631431945, | |
| "step": 29967 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_acc": 0.9871747501867955, | |
| "eval_f1": 0.9166171437939302, | |
| "eval_label.LOC.f1": 0.9542040545031572, | |
| "eval_label.LOC.p": 0.9512324410283594, | |
| "eval_label.LOC.r": 0.957194292572343, | |
| "eval_label.MISC.f1": 0.8256130790190737, | |
| "eval_label.MISC.p": 0.7881502890173411, | |
| "eval_label.MISC.r": 0.8668150031786396, | |
| "eval_label.ORG.f1": 0.9035007610350075, | |
| "eval_label.ORG.p": 0.8930344516323153, | |
| "eval_label.ORG.r": 0.914215308794086, | |
| "eval_label.PER.f1": 0.9314677327241576, | |
| "eval_label.PER.p": 0.9257840215694622, | |
| "eval_label.PER.r": 0.9372216635540871, | |
| "eval_loss": 0.059339962899684906, | |
| "eval_p": 0.9054655870445344, | |
| "eval_r": 0.928046806921449, | |
| "eval_runtime": 32.612, | |
| "eval_samples_per_second": 553.661, | |
| "eval_steps_per_second": 17.325, | |
| "step": 29967 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.656637966632843, | |
| "learning_rate": 1.2766332844611218e-05, | |
| "loss": 0.02074824090149881, | |
| "step": 34248 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_acc": 0.987262331918341, | |
| "eval_f1": 0.9183782232607716, | |
| "eval_label.LOC.f1": 0.9538971807628523, | |
| "eval_label.LOC.p": 0.9490496304118268, | |
| "eval_label.LOC.r": 0.9587945059341245, | |
| "eval_label.MISC.f1": 0.8343888537048765, | |
| "eval_label.MISC.p": 0.831230283911672, | |
| "eval_label.MISC.r": 0.8375715193897012, | |
| "eval_label.ORG.f1": 0.8998067489222537, | |
| "eval_label.ORG.p": 0.8695589714121534, | |
| "eval_label.ORG.r": 0.932234714307716, | |
| "eval_label.PER.f1": 0.9359443368481457, | |
| "eval_label.PER.p": 0.9346704871060172, | |
| "eval_label.PER.r": 0.9372216635540871, | |
| "eval_loss": 0.06151953712105751, | |
| "eval_p": 0.907441163365334, | |
| "eval_r": 0.9295821403377733, | |
| "eval_runtime": 32.5433, | |
| "eval_samples_per_second": 554.83, | |
| "eval_steps_per_second": 17.361, | |
| "step": 34248 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.0659022331237793, | |
| "learning_rate": 1.1702492482791184e-05, | |
| "loss": 0.01647565521332906, | |
| "step": 38529 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_acc": 0.987234962627233, | |
| "eval_f1": 0.9183171361791118, | |
| "eval_label.LOC.f1": 0.9529481834425253, | |
| "eval_label.LOC.p": 0.9458749343142406, | |
| "eval_label.LOC.r": 0.9601280170689426, | |
| "eval_label.MISC.f1": 0.8366175329712956, | |
| "eval_label.MISC.p": 0.8172173385874507, | |
| "eval_label.MISC.r": 0.8569612205975843, | |
| "eval_label.ORG.f1": 0.9041033779188392, | |
| "eval_label.ORG.p": 0.8875370919881306, | |
| "eval_label.ORG.r": 0.9212998613891884, | |
| "eval_label.PER.f1": 0.9319985698963175, | |
| "eval_label.PER.p": 0.9278189066059226, | |
| "eval_label.PER.r": 0.9362160609107887, | |
| "eval_loss": 0.06663035601377487, | |
| "eval_p": 0.9075987841945289, | |
| "eval_r": 0.9292916718536038, | |
| "eval_runtime": 32.3615, | |
| "eval_samples_per_second": 557.947, | |
| "eval_steps_per_second": 17.459, | |
| "step": 38529 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.06852112710475922, | |
| "learning_rate": 1.0638652120971149e-05, | |
| "loss": 0.012914503882126786, | |
| "step": 42810 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_acc": 0.9880013027782567, | |
| "eval_f1": 0.9225139285787711, | |
| "eval_label.LOC.f1": 0.9541126686061888, | |
| "eval_label.LOC.p": 0.9462295081967214, | |
| "eval_label.LOC.r": 0.9621282837711694, | |
| "eval_label.MISC.f1": 0.8479798773777708, | |
| "eval_label.MISC.p": 0.8388802488335926, | |
| "eval_label.MISC.r": 0.8572790845518118, | |
| "eval_label.ORG.f1": 0.9100626049778593, | |
| "eval_label.ORG.p": 0.9023467070401211, | |
| "eval_label.ORG.r": 0.9179115971045742, | |
| "eval_label.PER.f1": 0.933845497225779, | |
| "eval_label.PER.p": 0.924897844159504, | |
| "eval_label.PER.r": 0.942967964372935, | |
| "eval_loss": 0.07016361504793167, | |
| "eval_p": 0.9141879227446826, | |
| "eval_r": 0.9309929872608822, | |
| "eval_runtime": 32.9069, | |
| "eval_samples_per_second": 548.699, | |
| "eval_steps_per_second": 17.17, | |
| "step": 42810 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 1.4090399742126465, | |
| "learning_rate": 9.574811759151116e-06, | |
| "loss": 0.010248644444965309, | |
| "step": 47091 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_acc": 0.9879328795504868, | |
| "eval_f1": 0.9214825151863405, | |
| "eval_label.LOC.f1": 0.9554612937433723, | |
| "eval_label.LOC.p": 0.9497957570167348, | |
| "eval_label.LOC.r": 0.961194825976797, | |
| "eval_label.MISC.f1": 0.8431096511809792, | |
| "eval_label.MISC.p": 0.8299969202340622, | |
| "eval_label.MISC.r": 0.8566433566433567, | |
| "eval_label.ORG.f1": 0.9066000605510142, | |
| "eval_label.ORG.p": 0.8913528798928412, | |
| "eval_label.ORG.r": 0.9223779454797474, | |
| "eval_label.PER.f1": 0.9346633416458853, | |
| "eval_label.PER.p": 0.9271981905569692, | |
| "eval_label.PER.r": 0.9422496767705789, | |
| "eval_loss": 0.07261578738689423, | |
| "eval_p": 0.911567664135775, | |
| "eval_r": 0.9316154197269596, | |
| "eval_runtime": 32.2895, | |
| "eval_samples_per_second": 559.191, | |
| "eval_steps_per_second": 17.498, | |
| "step": 47091 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 2.3597915172576904, | |
| "learning_rate": 8.510971397331081e-06, | |
| "loss": 0.008314984113750711, | |
| "step": 51372 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_acc": 0.9877330837253985, | |
| "eval_f1": 0.9202380952380953, | |
| "eval_label.LOC.f1": 0.9561705508474576, | |
| "eval_label.LOC.p": 0.9495069033530572, | |
| "eval_label.LOC.r": 0.9629283904520602, | |
| "eval_label.MISC.f1": 0.8422717924379053, | |
| "eval_label.MISC.p": 0.8384251968503937, | |
| "eval_label.MISC.r": 0.8461538461538461, | |
| "eval_label.ORG.f1": 0.9035845795446007, | |
| "eval_label.ORG.p": 0.882301144702084, | |
| "eval_label.ORG.r": 0.9259202217772986, | |
| "eval_label.PER.f1": 0.932513583071204, | |
| "eval_label.PER.p": 0.9281343389782268, | |
| "eval_label.PER.r": 0.9369343485131446, | |
| "eval_loss": 0.07839750498533249, | |
| "eval_p": 0.9104829210836278, | |
| "eval_r": 0.9302045728038508, | |
| "eval_runtime": 32.9029, | |
| "eval_samples_per_second": 548.767, | |
| "eval_steps_per_second": 17.172, | |
| "step": 51372 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.18180809915065765, | |
| "learning_rate": 7.4471310355110474e-06, | |
| "loss": 0.006686228676617521, | |
| "step": 55653 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_acc": 0.9875524464040857, | |
| "eval_f1": 0.9198900693219574, | |
| "eval_label.LOC.f1": 0.9554880464931977, | |
| "eval_label.LOC.p": 0.9464869815517467, | |
| "eval_label.LOC.r": 0.9646619549273237, | |
| "eval_label.MISC.f1": 0.8379507816127534, | |
| "eval_label.MISC.p": 0.8165912518853695, | |
| "eval_label.MISC.r": 0.8604577240940877, | |
| "eval_label.ORG.f1": 0.9057290087905425, | |
| "eval_label.ORG.p": 0.8915411009995524, | |
| "eval_label.ORG.r": 0.9203757893115663, | |
| "eval_label.PER.f1": 0.9325882942904218, | |
| "eval_label.PER.p": 0.93012289225493, | |
| "eval_label.PER.r": 0.9350668007470191, | |
| "eval_loss": 0.08158153295516968, | |
| "eval_p": 0.9094448274463685, | |
| "eval_r": 0.9305780322834972, | |
| "eval_runtime": 32.4013, | |
| "eval_samples_per_second": 557.261, | |
| "eval_steps_per_second": 17.438, | |
| "step": 55653 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.18500462174415588, | |
| "learning_rate": 6.383290673691012e-06, | |
| "loss": 0.005471784470933978, | |
| "step": 59934 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_acc": 0.9871911717614602, | |
| "eval_f1": 0.9195246875640237, | |
| "eval_label.LOC.f1": 0.9548378539690962, | |
| "eval_label.LOC.p": 0.9497361477572559, | |
| "eval_label.LOC.r": 0.9599946659554607, | |
| "eval_label.MISC.f1": 0.8372093023255813, | |
| "eval_label.MISC.p": 0.8171912832929782, | |
| "eval_label.MISC.r": 0.8582326764144946, | |
| "eval_label.ORG.f1": 0.90653782273894, | |
| "eval_label.ORG.p": 0.8890863320005923, | |
| "eval_label.ORG.r": 0.9246881256738025, | |
| "eval_label.PER.f1": 0.9316708229426433, | |
| "eval_label.PER.p": 0.9242295730845349, | |
| "eval_label.PER.r": 0.9392328688406838, | |
| "eval_loss": 0.08883294463157654, | |
| "eval_p": 0.9081380761604144, | |
| "eval_r": 0.9312004647495746, | |
| "eval_runtime": 32.3818, | |
| "eval_samples_per_second": 557.597, | |
| "eval_steps_per_second": 17.448, | |
| "step": 59934 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.22286078333854675, | |
| "learning_rate": 5.319450311870977e-06, | |
| "loss": 0.004570412318357322, | |
| "step": 64215 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_acc": 0.988053304431362, | |
| "eval_f1": 0.9238630978137686, | |
| "eval_label.LOC.f1": 0.9564351175454908, | |
| "eval_label.LOC.p": 0.9526392379944437, | |
| "eval_label.LOC.r": 0.9602613681824244, | |
| "eval_label.MISC.f1": 0.8461170467230107, | |
| "eval_label.MISC.p": 0.84896, | |
| "eval_label.MISC.r": 0.8432930705657978, | |
| "eval_label.ORG.f1": 0.9100128661167033, | |
| "eval_label.ORG.p": 0.8946428571428572, | |
| "eval_label.ORG.r": 0.9259202217772986, | |
| "eval_label.PER.f1": 0.9367360068625348, | |
| "eval_label.PER.p": 0.9322709163346613, | |
| "eval_label.PER.r": 0.9412440741272806, | |
| "eval_loss": 0.0887371227145195, | |
| "eval_p": 0.9175671250818599, | |
| "eval_r": 0.9302460683015893, | |
| "eval_runtime": 32.8226, | |
| "eval_samples_per_second": 550.109, | |
| "eval_steps_per_second": 17.214, | |
| "step": 64215 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.09178588539361954, | |
| "learning_rate": 4.255609950050943e-06, | |
| "loss": 0.003622644723394546, | |
| "step": 68496 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_acc": 0.9879684596289271, | |
| "eval_f1": 0.9229315974289998, | |
| "eval_label.LOC.f1": 0.9551214361140442, | |
| "eval_label.LOC.p": 0.9455115640925127, | |
| "eval_label.LOC.r": 0.9649286571542872, | |
| "eval_label.MISC.f1": 0.8459366175086288, | |
| "eval_label.MISC.p": 0.8351920693928129, | |
| "eval_label.MISC.r": 0.8569612205975843, | |
| "eval_label.ORG.f1": 0.9098784194528875, | |
| "eval_label.ORG.p": 0.8980050997450127, | |
| "eval_label.ORG.r": 0.9220699214538735, | |
| "eval_label.PER.f1": 0.9354078937977304, | |
| "eval_label.PER.p": 0.9295035460992908, | |
| "eval_label.PER.r": 0.9413877316477518, | |
| "eval_loss": 0.09383999556303024, | |
| "eval_p": 0.913570208960078, | |
| "eval_r": 0.932486825179468, | |
| "eval_runtime": 32.431, | |
| "eval_samples_per_second": 556.751, | |
| "eval_steps_per_second": 17.422, | |
| "step": 68496 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.5162076354026794, | |
| "learning_rate": 3.191769588230909e-06, | |
| "loss": 0.002951568544704595, | |
| "step": 72777 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_acc": 0.9879985658491459, | |
| "eval_f1": 0.9234442866259083, | |
| "eval_label.LOC.f1": 0.9560388708931051, | |
| "eval_label.LOC.p": 0.947954902988988, | |
| "eval_label.LOC.r": 0.9642619015868783, | |
| "eval_label.MISC.f1": 0.8496981251986019, | |
| "eval_label.MISC.p": 0.849428208386277, | |
| "eval_label.MISC.r": 0.8499682136045772, | |
| "eval_label.ORG.f1": 0.910155952833777, | |
| "eval_label.ORG.p": 0.8992784125075165, | |
| "eval_label.ORG.r": 0.9212998613891884, | |
| "eval_label.PER.f1": 0.9338471419396276, | |
| "eval_label.PER.p": 0.9278218944980148, | |
| "eval_label.PER.r": 0.9399511564430398, | |
| "eval_loss": 0.09467604756355286, | |
| "eval_p": 0.9162581699346405, | |
| "eval_r": 0.9307440142744512, | |
| "eval_runtime": 32.3822, | |
| "eval_samples_per_second": 557.59, | |
| "eval_steps_per_second": 17.448, | |
| "step": 72777 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.003633791347965598, | |
| "learning_rate": 2.127929226410875e-06, | |
| "loss": 0.0021125146981239096, | |
| "step": 77058 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_acc": 0.9878316131733872, | |
| "eval_f1": 0.9223450284003527, | |
| "eval_label.LOC.f1": 0.9541914191419141, | |
| "eval_label.LOC.p": 0.9447131093974643, | |
| "eval_label.LOC.r": 0.9638618482464328, | |
| "eval_label.MISC.f1": 0.8473520249221183, | |
| "eval_label.MISC.p": 0.8307880268784361, | |
| "eval_label.MISC.r": 0.8645899554990464, | |
| "eval_label.ORG.f1": 0.9091460640072804, | |
| "eval_label.ORG.p": 0.8955625280143433, | |
| "eval_label.ORG.r": 0.9231480055444324, | |
| "eval_label.PER.f1": 0.9346941688673185, | |
| "eval_label.PER.p": 0.9287943262411348, | |
| "eval_label.PER.r": 0.9406694440453958, | |
| "eval_loss": 0.09958989918231964, | |
| "eval_p": 0.9117074752716069, | |
| "eval_r": 0.9332337441387609, | |
| "eval_runtime": 32.3603, | |
| "eval_samples_per_second": 557.968, | |
| "eval_steps_per_second": 17.46, | |
| "step": 77058 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.0006278403452597558, | |
| "learning_rate": 1.0640888645908402e-06, | |
| "loss": 0.0016824499075447836, | |
| "step": 81339 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_acc": 0.9881053060844671, | |
| "eval_f1": 0.92369345855854, | |
| "eval_label.LOC.f1": 0.9557276156442328, | |
| "eval_label.LOC.p": 0.9486337362059906, | |
| "eval_label.LOC.r": 0.9629283904520602, | |
| "eval_label.MISC.f1": 0.8502278799308502, | |
| "eval_label.MISC.p": 0.8408455082374884, | |
| "eval_label.MISC.r": 0.8598219961856326, | |
| "eval_label.ORG.f1": 0.911369332421964, | |
| "eval_label.ORG.p": 0.8990110878034162, | |
| "eval_label.ORG.r": 0.9240720776220546, | |
| "eval_label.PER.f1": 0.9340847481809103, | |
| "eval_label.PER.p": 0.9277313305937367, | |
| "eval_label.PER.r": 0.9405257865249246, | |
| "eval_loss": 0.09979883581399918, | |
| "eval_p": 0.9150244299674267, | |
| "eval_r": 0.9325283206772065, | |
| "eval_runtime": 32.7242, | |
| "eval_samples_per_second": 551.763, | |
| "eval_steps_per_second": 17.265, | |
| "step": 81339 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.0003999824111815542, | |
| "learning_rate": 2.485027708058945e-10, | |
| "loss": 0.0014637853742286275, | |
| "step": 85620 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_acc": 0.9881107799426887, | |
| "eval_f1": 0.9238326208484051, | |
| "eval_label.LOC.f1": 0.9561130601707817, | |
| "eval_label.LOC.p": 0.9492639327024185, | |
| "eval_label.LOC.r": 0.963061741565542, | |
| "eval_label.MISC.f1": 0.8494572911750827, | |
| "eval_label.MISC.p": 0.8408595453129866, | |
| "eval_label.MISC.r": 0.8582326764144946, | |
| "eval_label.ORG.f1": 0.9107603581727122, | |
| "eval_label.ORG.p": 0.8976813762154077, | |
| "eval_label.ORG.r": 0.9242260896349915, | |
| "eval_label.PER.f1": 0.935064935064935, | |
| "eval_label.PER.p": 0.9289663972777541, | |
| "eval_label.PER.r": 0.9412440741272806, | |
| "eval_loss": 0.10018600523471832, | |
| "eval_p": 0.9152176568799121, | |
| "eval_r": 0.9326113116726835, | |
| "eval_runtime": 32.2631, | |
| "eval_samples_per_second": 559.649, | |
| "eval_steps_per_second": 17.512, | |
| "step": 85620 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 85620, | |
| "total_flos": 1.3381069410175074e+17, | |
| "train_loss": 0.040764350686165754, | |
| "train_runtime": 15630.4696, | |
| "train_samples_per_second": 175.264, | |
| "train_steps_per_second": 5.478 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 85620, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3381069410175074e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |