Instructions to use redqueen01/french-summariser with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use redqueen01/french-summariser with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("redqueen01/french-summariser") model = AutoModelForSeq2SeqLM.from_pretrained("redqueen01/french-summariser", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download trainer_state.json from redqueen01/french-summariser: direct link, hf CLI and curl.
- Browser
- Download file 28.1 kB
-
https://huggingface.co/redqueen01/french-summariser/resolve/main/trainer_state.json
- Command line
-
hf download hf://redqueen01/french-summariser/trainer_state.json
-
curl -L -o trainer_state.json https://huggingface.co/redqueen01/french-summariser/resolve/main/trainer_state.json
28.1 kB
{
"best_metric": 36.6877,
"best_model_checkpoint": "/opt/ml/model/checkpoint-22098",
"epoch": 4.0,
"eval_steps": 500,
"global_step": 29464,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.027151778441487917,
"grad_norm": 37.803707122802734,
"learning_rate": 2.7137042062415198e-06,
"loss": 5.331,
"step": 200
},
{
"epoch": 0.054303556882975834,
"grad_norm": 23.336469650268555,
"learning_rate": 5.4274084124830395e-06,
"loss": 4.4602,
"step": 400
},
{
"epoch": 0.08145533532446375,
"grad_norm": 22.160137176513672,
"learning_rate": 8.14111261872456e-06,
"loss": 3.8433,
"step": 600
},
{
"epoch": 0.10860711376595167,
"grad_norm": 29.797374725341797,
"learning_rate": 1.0854816824966079e-05,
"loss": 2.8494,
"step": 800
},
{
"epoch": 0.13575889220743959,
"grad_norm": 12.725276947021484,
"learning_rate": 1.35685210312076e-05,
"loss": 2.3127,
"step": 1000
},
{
"epoch": 0.1629106706489275,
"grad_norm": 14.782560348510742,
"learning_rate": 1.628222523744912e-05,
"loss": 2.1081,
"step": 1200
},
{
"epoch": 0.19006244909041542,
"grad_norm": 10.947192192077637,
"learning_rate": 1.899592944369064e-05,
"loss": 2.0052,
"step": 1400
},
{
"epoch": 0.21721422753190334,
"grad_norm": 13.707736015319824,
"learning_rate": 1.9909967845659165e-05,
"loss": 1.9766,
"step": 1600
},
{
"epoch": 0.24436600597339125,
"grad_norm": 8.410202980041504,
"learning_rate": 1.9767059664165775e-05,
"loss": 1.8809,
"step": 1800
},
{
"epoch": 0.27151778441487917,
"grad_norm": 24.935426712036133,
"learning_rate": 1.9624151482672384e-05,
"loss": 2.4196,
"step": 2000
},
{
"epoch": 0.2986695628563671,
"grad_norm": 13.886157989501953,
"learning_rate": 1.9481243301178994e-05,
"loss": 1.8704,
"step": 2200
},
{
"epoch": 0.325821341297855,
"grad_norm": 7.822121620178223,
"learning_rate": 1.9338335119685603e-05,
"loss": 1.8206,
"step": 2400
},
{
"epoch": 0.3529731197393429,
"grad_norm": 7.880593299865723,
"learning_rate": 1.9195426938192212e-05,
"loss": 1.9249,
"step": 2600
},
{
"epoch": 0.38012489818083084,
"grad_norm": 8.565655708312988,
"learning_rate": 1.9052518756698825e-05,
"loss": 1.8616,
"step": 2800
},
{
"epoch": 0.40727667662231876,
"grad_norm": 11.997269630432129,
"learning_rate": 1.890961057520543e-05,
"loss": 1.96,
"step": 3000
},
{
"epoch": 0.4344284550638067,
"grad_norm": 7.973113059997559,
"learning_rate": 1.876670239371204e-05,
"loss": 1.8759,
"step": 3200
},
{
"epoch": 0.4615802335052946,
"grad_norm": 8.130441665649414,
"learning_rate": 1.862379421221865e-05,
"loss": 1.7713,
"step": 3400
},
{
"epoch": 0.4887320119467825,
"grad_norm": 14.352017402648926,
"learning_rate": 1.848088603072526e-05,
"loss": 1.8688,
"step": 3600
},
{
"epoch": 0.5158837903882705,
"grad_norm": 12.24020767211914,
"learning_rate": 1.833797784923187e-05,
"loss": 1.7597,
"step": 3800
},
{
"epoch": 0.5430355688297583,
"grad_norm": 9.355463027954102,
"learning_rate": 1.8195069667738478e-05,
"loss": 1.7855,
"step": 4000
},
{
"epoch": 0.5701873472712463,
"grad_norm": 9.409703254699707,
"learning_rate": 1.805216148624509e-05,
"loss": 1.8071,
"step": 4200
},
{
"epoch": 0.5973391257127342,
"grad_norm": 12.054621696472168,
"learning_rate": 1.79092533047517e-05,
"loss": 1.7745,
"step": 4400
},
{
"epoch": 0.6244909041542221,
"grad_norm": 7.502680778503418,
"learning_rate": 1.776634512325831e-05,
"loss": 1.82,
"step": 4600
},
{
"epoch": 0.65164268259571,
"grad_norm": 11.708473205566406,
"learning_rate": 1.762343694176492e-05,
"loss": 1.751,
"step": 4800
},
{
"epoch": 0.678794461037198,
"grad_norm": 7.169336795806885,
"learning_rate": 1.7480528760271528e-05,
"loss": 1.7742,
"step": 5000
},
{
"epoch": 0.7059462394786858,
"grad_norm": 12.792072296142578,
"learning_rate": 1.7337620578778138e-05,
"loss": 1.6788,
"step": 5200
},
{
"epoch": 0.7330980179201738,
"grad_norm": 6.759461879730225,
"learning_rate": 1.7194712397284744e-05,
"loss": 1.7235,
"step": 5400
},
{
"epoch": 0.7602497963616617,
"grad_norm": 9.167635917663574,
"learning_rate": 1.7051804215791353e-05,
"loss": 2.0984,
"step": 5600
},
{
"epoch": 0.7874015748031497,
"grad_norm": 15.99834156036377,
"learning_rate": 1.6908896034297966e-05,
"loss": 1.8718,
"step": 5800
},
{
"epoch": 0.8145533532446375,
"grad_norm": 8.619094848632812,
"learning_rate": 1.6765987852804575e-05,
"loss": 1.6876,
"step": 6000
},
{
"epoch": 0.8417051316861255,
"grad_norm": 7.159388065338135,
"learning_rate": 1.6623079671311184e-05,
"loss": 1.6663,
"step": 6200
},
{
"epoch": 0.8688569101276133,
"grad_norm": 7.572915077209473,
"learning_rate": 1.6480171489817794e-05,
"loss": 1.7924,
"step": 6400
},
{
"epoch": 0.8960086885691013,
"grad_norm": 9.616636276245117,
"learning_rate": 1.6337263308324403e-05,
"loss": 1.7829,
"step": 6600
},
{
"epoch": 0.9231604670105892,
"grad_norm": 14.096186637878418,
"learning_rate": 1.6194355126831013e-05,
"loss": 1.7055,
"step": 6800
},
{
"epoch": 0.9503122454520772,
"grad_norm": 10.016873359680176,
"learning_rate": 1.6051446945337622e-05,
"loss": 1.723,
"step": 7000
},
{
"epoch": 0.977464023893565,
"grad_norm": 11.668209075927734,
"learning_rate": 1.590853876384423e-05,
"loss": 1.8303,
"step": 7200
},
{
"epoch": 1.0,
"eval_gen_len": 20.185819070904646,
"eval_loss": 1.6442582607269287,
"eval_rouge1": 39.7314,
"eval_rouge2": 18.4816,
"eval_rougeL": 33.172,
"eval_rougeLsum": 34.4741,
"eval_runtime": 225.7579,
"eval_samples_per_second": 3.623,
"eval_steps_per_second": 3.623,
"step": 7366
},
{
"epoch": 1.0046158023350529,
"grad_norm": 7.712486267089844,
"learning_rate": 1.576563058235084e-05,
"loss": 1.737,
"step": 7400
},
{
"epoch": 1.031767580776541,
"grad_norm": 10.486478805541992,
"learning_rate": 1.562272240085745e-05,
"loss": 1.4298,
"step": 7600
},
{
"epoch": 1.0589193592180288,
"grad_norm": 7.863742351531982,
"learning_rate": 1.547981421936406e-05,
"loss": 1.3783,
"step": 7800
},
{
"epoch": 1.0860711376595167,
"grad_norm": 7.61310338973999,
"learning_rate": 1.533690603787067e-05,
"loss": 1.3183,
"step": 8000
},
{
"epoch": 1.1132229161010045,
"grad_norm": 8.521239280700684,
"learning_rate": 1.5193997856377278e-05,
"loss": 1.3318,
"step": 8200
},
{
"epoch": 1.1403746945424926,
"grad_norm": 11.96485424041748,
"learning_rate": 1.5051089674883888e-05,
"loss": 1.4572,
"step": 8400
},
{
"epoch": 1.1675264729839805,
"grad_norm": 7.318172931671143,
"learning_rate": 1.4908181493390499e-05,
"loss": 1.3739,
"step": 8600
},
{
"epoch": 1.1946782514254684,
"grad_norm": 8.905569076538086,
"learning_rate": 1.4765273311897108e-05,
"loss": 1.4641,
"step": 8800
},
{
"epoch": 1.2218300298669562,
"grad_norm": 14.515080451965332,
"learning_rate": 1.4622365130403717e-05,
"loss": 1.3409,
"step": 9000
},
{
"epoch": 1.2489818083084443,
"grad_norm": 13.09267807006836,
"learning_rate": 1.4479456948910327e-05,
"loss": 1.3698,
"step": 9200
},
{
"epoch": 1.2761335867499322,
"grad_norm": 7.095240116119385,
"learning_rate": 1.4336548767416936e-05,
"loss": 1.3622,
"step": 9400
},
{
"epoch": 1.30328536519142,
"grad_norm": 8.748358726501465,
"learning_rate": 1.4193640585923546e-05,
"loss": 1.3283,
"step": 9600
},
{
"epoch": 1.3304371436329079,
"grad_norm": 9.457386016845703,
"learning_rate": 1.4050732404430155e-05,
"loss": 1.3209,
"step": 9800
},
{
"epoch": 1.3575889220743957,
"grad_norm": 8.674896240234375,
"learning_rate": 1.3907824222936763e-05,
"loss": 1.3939,
"step": 10000
},
{
"epoch": 1.3847407005158838,
"grad_norm": 8.862044334411621,
"learning_rate": 1.3764916041443375e-05,
"loss": 1.3447,
"step": 10200
},
{
"epoch": 1.4118924789573717,
"grad_norm": 8.570720672607422,
"learning_rate": 1.3622007859949983e-05,
"loss": 1.4332,
"step": 10400
},
{
"epoch": 1.4390442573988595,
"grad_norm": 7.600680828094482,
"learning_rate": 1.3479099678456592e-05,
"loss": 1.3661,
"step": 10600
},
{
"epoch": 1.4661960358403476,
"grad_norm": 12.70207691192627,
"learning_rate": 1.3336191496963202e-05,
"loss": 1.328,
"step": 10800
},
{
"epoch": 1.4933478142818355,
"grad_norm": 7.30202579498291,
"learning_rate": 1.3193283315469811e-05,
"loss": 1.3193,
"step": 11000
},
{
"epoch": 1.5204995927233234,
"grad_norm": 9.046040534973145,
"learning_rate": 1.305037513397642e-05,
"loss": 1.3822,
"step": 11200
},
{
"epoch": 1.5476513711648114,
"grad_norm": 15.81724739074707,
"learning_rate": 1.290746695248303e-05,
"loss": 6.8428,
"step": 11400
},
{
"epoch": 1.574803149606299,
"grad_norm": 7.630992412567139,
"learning_rate": 1.2764558770989641e-05,
"loss": 1.5811,
"step": 11600
},
{
"epoch": 1.6019549280477872,
"grad_norm": 7.0756611824035645,
"learning_rate": 1.262165058949625e-05,
"loss": 1.3155,
"step": 11800
},
{
"epoch": 1.629106706489275,
"grad_norm": 4.655889987945557,
"learning_rate": 1.247874240800286e-05,
"loss": 1.3445,
"step": 12000
},
{
"epoch": 1.6562584849307629,
"grad_norm": 10.755663871765137,
"learning_rate": 1.2335834226509469e-05,
"loss": 1.3034,
"step": 12200
},
{
"epoch": 1.683410263372251,
"grad_norm": 9.8883056640625,
"learning_rate": 1.2192926045016078e-05,
"loss": 1.3155,
"step": 12400
},
{
"epoch": 1.7105620418137388,
"grad_norm": 9.121843338012695,
"learning_rate": 1.2050017863522688e-05,
"loss": 1.3521,
"step": 12600
},
{
"epoch": 1.7377138202552267,
"grad_norm": 8.46533489227295,
"learning_rate": 1.1907109682029297e-05,
"loss": 1.2941,
"step": 12800
},
{
"epoch": 1.7648655986967148,
"grad_norm": 10.633326530456543,
"learning_rate": 1.1764201500535905e-05,
"loss": 1.3283,
"step": 13000
},
{
"epoch": 1.7920173771382024,
"grad_norm": 15.54390811920166,
"learning_rate": 1.1621293319042518e-05,
"loss": 1.3425,
"step": 13200
},
{
"epoch": 1.8191691555796905,
"grad_norm": 8.352429389953613,
"learning_rate": 1.1478385137549125e-05,
"loss": 1.2978,
"step": 13400
},
{
"epoch": 1.8463209340211784,
"grad_norm": 9.304736137390137,
"learning_rate": 1.1335476956055735e-05,
"loss": 1.3825,
"step": 13600
},
{
"epoch": 1.8734727124626662,
"grad_norm": 7.9684062004089355,
"learning_rate": 1.1192568774562344e-05,
"loss": 1.3417,
"step": 13800
},
{
"epoch": 1.9006244909041543,
"grad_norm": 7.230633735656738,
"learning_rate": 1.1049660593068954e-05,
"loss": 1.2805,
"step": 14000
},
{
"epoch": 1.9277762693456422,
"grad_norm": 10.64439582824707,
"learning_rate": 1.0906752411575563e-05,
"loss": 1.3003,
"step": 14200
},
{
"epoch": 1.95492804778713,
"grad_norm": 8.245725631713867,
"learning_rate": 1.0763844230082172e-05,
"loss": 1.347,
"step": 14400
},
{
"epoch": 1.9820798262286181,
"grad_norm": 8.880694389343262,
"learning_rate": 1.0620936048588783e-05,
"loss": 1.2425,
"step": 14600
},
{
"epoch": 2.0,
"eval_gen_len": 21.359413202933986,
"eval_loss": 1.5719200372695923,
"eval_rouge1": 41.4083,
"eval_rouge2": 20.3324,
"eval_rougeL": 35.0169,
"eval_rougeLsum": 36.1426,
"eval_runtime": 240.7643,
"eval_samples_per_second": 3.398,
"eval_steps_per_second": 3.398,
"step": 14732
},
{
"epoch": 2.0092316046701058,
"grad_norm": 9.85111141204834,
"learning_rate": 1.0478027867095393e-05,
"loss": 1.2135,
"step": 14800
},
{
"epoch": 2.036383383111594,
"grad_norm": 10.956082344055176,
"learning_rate": 1.0335119685602002e-05,
"loss": 0.9318,
"step": 15000
},
{
"epoch": 2.063535161553082,
"grad_norm": 9.48881721496582,
"learning_rate": 1.0192211504108611e-05,
"loss": 0.9307,
"step": 15200
},
{
"epoch": 2.0906869399945696,
"grad_norm": 7.347820281982422,
"learning_rate": 1.004930332261522e-05,
"loss": 0.9626,
"step": 15400
},
{
"epoch": 2.1178387184360576,
"grad_norm": 12.733508110046387,
"learning_rate": 9.90639514112183e-06,
"loss": 0.9948,
"step": 15600
},
{
"epoch": 2.1449904968775453,
"grad_norm": 10.213859558105469,
"learning_rate": 9.76348695962844e-06,
"loss": 0.9832,
"step": 15800
},
{
"epoch": 2.1721422753190334,
"grad_norm": 11.045472145080566,
"learning_rate": 9.620578778135049e-06,
"loss": 0.9583,
"step": 16000
},
{
"epoch": 2.1992940537605215,
"grad_norm": 6.748650074005127,
"learning_rate": 9.477670596641658e-06,
"loss": 0.9597,
"step": 16200
},
{
"epoch": 2.226445832202009,
"grad_norm": 9.791770935058594,
"learning_rate": 9.334762415148268e-06,
"loss": 0.9562,
"step": 16400
},
{
"epoch": 2.253597610643497,
"grad_norm": 9.312710762023926,
"learning_rate": 9.191854233654877e-06,
"loss": 0.9233,
"step": 16600
},
{
"epoch": 2.2807493890849853,
"grad_norm": 7.41344690322876,
"learning_rate": 9.048946052161486e-06,
"loss": 0.8873,
"step": 16800
},
{
"epoch": 2.307901167526473,
"grad_norm": 7.966084003448486,
"learning_rate": 8.906037870668096e-06,
"loss": 0.9414,
"step": 17000
},
{
"epoch": 2.335052945967961,
"grad_norm": 13.925320625305176,
"learning_rate": 8.763129689174707e-06,
"loss": 0.9562,
"step": 17200
},
{
"epoch": 2.362204724409449,
"grad_norm": 6.855761528015137,
"learning_rate": 8.620221507681316e-06,
"loss": 0.9708,
"step": 17400
},
{
"epoch": 2.3893565028509367,
"grad_norm": 7.2677106857299805,
"learning_rate": 8.477313326187924e-06,
"loss": 1.0446,
"step": 17600
},
{
"epoch": 2.416508281292425,
"grad_norm": 10.490965843200684,
"learning_rate": 8.334405144694533e-06,
"loss": 0.9229,
"step": 17800
},
{
"epoch": 2.4436600597339124,
"grad_norm": 11.04757308959961,
"learning_rate": 8.191496963201144e-06,
"loss": 0.9562,
"step": 18000
},
{
"epoch": 2.4708118381754005,
"grad_norm": 10.986570358276367,
"learning_rate": 8.048588781707754e-06,
"loss": 0.9926,
"step": 18200
},
{
"epoch": 2.4979636166168886,
"grad_norm": 7.6767988204956055,
"learning_rate": 7.905680600214363e-06,
"loss": 0.9538,
"step": 18400
},
{
"epoch": 2.5251153950583762,
"grad_norm": 10.37006950378418,
"learning_rate": 7.762772418720973e-06,
"loss": 0.9283,
"step": 18600
},
{
"epoch": 2.5522671734998643,
"grad_norm": 6.103606700897217,
"learning_rate": 7.619864237227582e-06,
"loss": 0.9557,
"step": 18800
},
{
"epoch": 2.579418951941352,
"grad_norm": 10.271400451660156,
"learning_rate": 7.476956055734191e-06,
"loss": 0.992,
"step": 19000
},
{
"epoch": 2.60657073038284,
"grad_norm": 9.806522369384766,
"learning_rate": 7.334047874240801e-06,
"loss": 0.9976,
"step": 19200
},
{
"epoch": 2.633722508824328,
"grad_norm": 8.216267585754395,
"learning_rate": 7.191139692747411e-06,
"loss": 1.0236,
"step": 19400
},
{
"epoch": 2.6608742872658158,
"grad_norm": 7.928046226501465,
"learning_rate": 7.04823151125402e-06,
"loss": 0.9992,
"step": 19600
},
{
"epoch": 2.688026065707304,
"grad_norm": 12.21468734741211,
"learning_rate": 6.90532332976063e-06,
"loss": 0.9522,
"step": 19800
},
{
"epoch": 2.7151778441487915,
"grad_norm": 10.603358268737793,
"learning_rate": 6.762415148267238e-06,
"loss": 0.9974,
"step": 20000
},
{
"epoch": 2.7423296225902796,
"grad_norm": 7.231788158416748,
"learning_rate": 6.619506966773848e-06,
"loss": 1.0106,
"step": 20200
},
{
"epoch": 2.7694814010317677,
"grad_norm": 10.388824462890625,
"learning_rate": 6.476598785280458e-06,
"loss": 0.9749,
"step": 20400
},
{
"epoch": 2.7966331794732557,
"grad_norm": 15.31623649597168,
"learning_rate": 6.333690603787067e-06,
"loss": 1.0069,
"step": 20600
},
{
"epoch": 2.8237849579147434,
"grad_norm": 8.504380226135254,
"learning_rate": 6.1907824222936765e-06,
"loss": 0.9933,
"step": 20800
},
{
"epoch": 2.8509367363562315,
"grad_norm": 9.712935447692871,
"learning_rate": 6.047874240800287e-06,
"loss": 0.9548,
"step": 21000
},
{
"epoch": 2.878088514797719,
"grad_norm": 7.156877517700195,
"learning_rate": 5.904966059306896e-06,
"loss": 0.9425,
"step": 21200
},
{
"epoch": 2.905240293239207,
"grad_norm": 15.062908172607422,
"learning_rate": 5.762057877813505e-06,
"loss": 0.9668,
"step": 21400
},
{
"epoch": 2.9323920716806953,
"grad_norm": 12.548048973083496,
"learning_rate": 5.619149696320115e-06,
"loss": 0.9691,
"step": 21600
},
{
"epoch": 2.959543850122183,
"grad_norm": 10.592756271362305,
"learning_rate": 5.476241514826724e-06,
"loss": 0.9379,
"step": 21800
},
{
"epoch": 2.986695628563671,
"grad_norm": 7.80446720123291,
"learning_rate": 5.333333333333334e-06,
"loss": 0.9629,
"step": 22000
},
{
"epoch": 3.0,
"eval_gen_len": 22.09046454767726,
"eval_loss": 1.648707628250122,
"eval_rouge1": 42.3336,
"eval_rouge2": 20.7191,
"eval_rougeL": 35.3723,
"eval_rougeLsum": 36.6877,
"eval_runtime": 249.7303,
"eval_samples_per_second": 3.276,
"eval_steps_per_second": 3.276,
"step": 22098
},
{
"epoch": 3.0138474070051586,
"grad_norm": 8.314491271972656,
"learning_rate": 5.190425151839943e-06,
"loss": 0.7822,
"step": 22200
},
{
"epoch": 3.0409991854466467,
"grad_norm": 9.839071273803711,
"learning_rate": 5.047516970346553e-06,
"loss": 0.7267,
"step": 22400
},
{
"epoch": 3.068150963888135,
"grad_norm": 7.027435779571533,
"learning_rate": 4.904608788853163e-06,
"loss": 0.6844,
"step": 22600
},
{
"epoch": 3.0953027423296224,
"grad_norm": 9.001220703125,
"learning_rate": 4.761700607359771e-06,
"loss": 0.7271,
"step": 22800
},
{
"epoch": 3.1224545207711105,
"grad_norm": 13.049047470092773,
"learning_rate": 4.618792425866381e-06,
"loss": 0.7248,
"step": 23000
},
{
"epoch": 3.1496062992125986,
"grad_norm": 8.460586547851562,
"learning_rate": 4.475884244372991e-06,
"loss": 0.7289,
"step": 23200
},
{
"epoch": 3.1767580776540862,
"grad_norm": 8.50259780883789,
"learning_rate": 4.3329760628796e-06,
"loss": 0.7217,
"step": 23400
},
{
"epoch": 3.2039098560955743,
"grad_norm": 8.076554298400879,
"learning_rate": 4.1900678813862095e-06,
"loss": 0.7232,
"step": 23600
},
{
"epoch": 3.2310616345370624,
"grad_norm": 11.854021072387695,
"learning_rate": 4.047159699892819e-06,
"loss": 0.698,
"step": 23800
},
{
"epoch": 3.25821341297855,
"grad_norm": 10.784579277038574,
"learning_rate": 3.904251518399429e-06,
"loss": 0.6738,
"step": 24000
},
{
"epoch": 3.285365191420038,
"grad_norm": 9.391136169433594,
"learning_rate": 3.761343336906038e-06,
"loss": 0.6969,
"step": 24200
},
{
"epoch": 3.3125169698615258,
"grad_norm": 7.319575786590576,
"learning_rate": 3.618435155412648e-06,
"loss": 0.6977,
"step": 24400
},
{
"epoch": 3.339668748303014,
"grad_norm": 7.981252193450928,
"learning_rate": 3.4755269739192572e-06,
"loss": 0.7155,
"step": 24600
},
{
"epoch": 3.366820526744502,
"grad_norm": 9.300016403198242,
"learning_rate": 3.3326187924258666e-06,
"loss": 0.72,
"step": 24800
},
{
"epoch": 3.3939723051859896,
"grad_norm": 9.198232650756836,
"learning_rate": 3.189710610932476e-06,
"loss": 0.6914,
"step": 25000
},
{
"epoch": 3.4211240836274777,
"grad_norm": 9.116570472717285,
"learning_rate": 3.046802429439086e-06,
"loss": 0.6684,
"step": 25200
},
{
"epoch": 3.4482758620689653,
"grad_norm": 10.804572105407715,
"learning_rate": 2.9038942479456948e-06,
"loss": 0.7164,
"step": 25400
},
{
"epoch": 3.4754276405104534,
"grad_norm": 8.90661907196045,
"learning_rate": 2.7609860664523046e-06,
"loss": 0.6729,
"step": 25600
},
{
"epoch": 3.5025794189519415,
"grad_norm": 13.364439964294434,
"learning_rate": 2.618077884958914e-06,
"loss": 0.7134,
"step": 25800
},
{
"epoch": 3.529731197393429,
"grad_norm": 9.653249740600586,
"learning_rate": 2.4751697034655237e-06,
"loss": 0.6927,
"step": 26000
},
{
"epoch": 3.556882975834917,
"grad_norm": 7.737614154815674,
"learning_rate": 2.332261521972133e-06,
"loss": 0.6489,
"step": 26200
},
{
"epoch": 3.584034754276405,
"grad_norm": 10.83629322052002,
"learning_rate": 2.1893533404787425e-06,
"loss": 0.6121,
"step": 26400
},
{
"epoch": 3.611186532717893,
"grad_norm": 4.974267482757568,
"learning_rate": 2.046445158985352e-06,
"loss": 0.6844,
"step": 26600
},
{
"epoch": 3.638338311159381,
"grad_norm": 9.46348762512207,
"learning_rate": 1.9035369774919615e-06,
"loss": 0.6942,
"step": 26800
},
{
"epoch": 3.665490089600869,
"grad_norm": 10.359207153320312,
"learning_rate": 1.760628795998571e-06,
"loss": 0.7198,
"step": 27000
},
{
"epoch": 3.6926418680423567,
"grad_norm": 8.743695259094238,
"learning_rate": 1.6177206145051804e-06,
"loss": 0.6993,
"step": 27200
},
{
"epoch": 3.719793646483845,
"grad_norm": 14.00108814239502,
"learning_rate": 1.47481243301179e-06,
"loss": 0.658,
"step": 27400
},
{
"epoch": 3.7469454249253324,
"grad_norm": 8.488143920898438,
"learning_rate": 1.3319042515183994e-06,
"loss": 0.705,
"step": 27600
},
{
"epoch": 3.7740972033668205,
"grad_norm": 8.011665344238281,
"learning_rate": 1.188996070025009e-06,
"loss": 0.7185,
"step": 27800
},
{
"epoch": 3.8012489818083086,
"grad_norm": 15.173530578613281,
"learning_rate": 1.0460878885316186e-06,
"loss": 0.6464,
"step": 28000
},
{
"epoch": 3.8284007602497963,
"grad_norm": 10.643668174743652,
"learning_rate": 9.031797070382279e-07,
"loss": 0.7051,
"step": 28200
},
{
"epoch": 3.8555525386912843,
"grad_norm": 6.029738426208496,
"learning_rate": 7.602715255448374e-07,
"loss": 0.6912,
"step": 28400
},
{
"epoch": 3.882704317132772,
"grad_norm": 12.955596923828125,
"learning_rate": 6.17363344051447e-07,
"loss": 0.66,
"step": 28600
},
{
"epoch": 3.90985609557426,
"grad_norm": 8.502635955810547,
"learning_rate": 4.744551625580565e-07,
"loss": 0.7093,
"step": 28800
},
{
"epoch": 3.937007874015748,
"grad_norm": 6.528735160827637,
"learning_rate": 3.3154698106466597e-07,
"loss": 0.6797,
"step": 29000
},
{
"epoch": 3.964159652457236,
"grad_norm": 9.745908737182617,
"learning_rate": 1.8863879957127545e-07,
"loss": 0.7096,
"step": 29200
},
{
"epoch": 3.991311430898724,
"grad_norm": 6.438901424407959,
"learning_rate": 4.573061807788496e-08,
"loss": 0.685,
"step": 29400
},
{
"epoch": 4.0,
"eval_gen_len": 25.04156479217604,
"eval_loss": 1.814982295036316,
"eval_rouge1": 42.4421,
"eval_rouge2": 19.9827,
"eval_rougeL": 34.7409,
"eval_rougeLsum": 36.2234,
"eval_runtime": 275.069,
"eval_samples_per_second": 2.974,
"eval_steps_per_second": 2.974,
"step": 29464
},
{
"epoch": 4.0,
"step": 29464,
"total_flos": 1.4564347981725696e+16,
"train_loss": 1.318135852435669,
"train_runtime": 8629.8464,
"train_samples_per_second": 6.828,
"train_steps_per_second": 3.414
}
],
"logging_steps": 200,
"max_steps": 29464,
"num_input_tokens_seen": 0,
"num_train_epochs": 4,
"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": 1.4564347981725696e+16,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}
{
"best_metric": 36.6877,
"best_model_checkpoint": "/opt/ml/model/checkpoint-22098",
"epoch": 4.0,
"eval_steps": 500,
"global_step": 29464,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.027151778441487917,
"grad_norm": 37.803707122802734,
"learning_rate": 2.7137042062415198e-06,
"loss": 5.331,
"step": 200
},
{
"epoch": 0.054303556882975834,
"grad_norm": 23.336469650268555,
"learning_rate": 5.4274084124830395e-06,
"loss": 4.4602,
"step": 400
},
{
"epoch": 0.08145533532446375,
"grad_norm": 22.160137176513672,
"learning_rate": 8.14111261872456e-06,
"loss": 3.8433,
"step": 600
},
{
"epoch": 0.10860711376595167,
"grad_norm": 29.797374725341797,
"learning_rate": 1.0854816824966079e-05,
"loss": 2.8494,
"step": 800
},
{
"epoch": 0.13575889220743959,
"grad_norm": 12.725276947021484,
"learning_rate": 1.35685210312076e-05,
"loss": 2.3127,
"step": 1000
},
{
"epoch": 0.1629106706489275,
"grad_norm": 14.782560348510742,
"learning_rate": 1.628222523744912e-05,
"loss": 2.1081,
"step": 1200
},
{
"epoch": 0.19006244909041542,
"grad_norm": 10.947192192077637,
"learning_rate": 1.899592944369064e-05,
"loss": 2.0052,
"step": 1400
},
{
"epoch": 0.21721422753190334,
"grad_norm": 13.707736015319824,
"learning_rate": 1.9909967845659165e-05,
"loss": 1.9766,
"step": 1600
},
{
"epoch": 0.24436600597339125,
"grad_norm": 8.410202980041504,
"learning_rate": 1.9767059664165775e-05,
"loss": 1.8809,
"step": 1800
},
{
"epoch": 0.27151778441487917,
"grad_norm": 24.935426712036133,
"learning_rate": 1.9624151482672384e-05,
"loss": 2.4196,
"step": 2000
},
{
"epoch": 0.2986695628563671,
"grad_norm": 13.886157989501953,
"learning_rate": 1.9481243301178994e-05,
"loss": 1.8704,
"step": 2200
},
{
"epoch": 0.325821341297855,
"grad_norm": 7.822121620178223,
"learning_rate": 1.9338335119685603e-05,
"loss": 1.8206,
"step": 2400
},
{
"epoch": 0.3529731197393429,
"grad_norm": 7.880593299865723,
"learning_rate": 1.9195426938192212e-05,
"loss": 1.9249,
"step": 2600