CoolFace
Modelpublic

tsavage68/Transaminitis_L3_1000steps_1e5rate_01beta_CSFTDPO

sourceHugging Facellama3updated 2y agoView on Hugging Face
0likes14downloads
Model Card

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

TransaminitisL31000steps1e5rate01beta_CSFTDPO

This model is a fine-tuned version of tsavage68/Transaminitis_L3_1000rate_1e7_SFT on an unknown dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.2656
  • —Rewards/chosen: -5.8394
  • —Rewards/rejected: -13.5464
  • —Rewards/accuracies: 0.9500
  • —Rewards/margins: 7.7070
  • —Logps/rejected: -154.0191
  • —Logps/chosen: -76.9285
  • —Logits/rejected: -1.0971
  • —Logits/chosen: -1.0952

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • —learning_rate: 1e-05
  • —trainbatchsize: 2
  • —evalbatchsize: 1
  • —seed: 42
  • —gradientaccumulationsteps: 2
  • —totaltrainbatch_size: 4
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: cosine
  • —lrschedulerwarmup_steps: 100
  • —training_steps: 1000

Training results

Training LossEpochStepValidation LossRewards/chosenRewards/rejectedRewards/accuraciesRewards/marginsLogps/rejectedLogps/chosenLogits/rejectedLogits/chosen
0.81570.2250.7130-1.6812-1.64220.2000-0.0390-34.9765-35.3459-1.0074-1.0078
0.65310.4500.55721.49201.09990.54000.3921-7.5562-3.6147-0.6331-0.6288
0.00690.6750.06381.5026-8.01720.99009.5198-98.7265-3.5080-1.0032-0.9076
1.49870.81000.7768-3.4746-3.53220.54000.0576-53.8765-53.2803-0.4138-0.4136
0.79871.01250.7220-3.4829-3.51100.54000.0281-53.6649-53.3632-0.7087-0.7087
0.74381.21500.7114-3.2843-3.25350.4600-0.0308-51.0900-51.3775-1.0310-1.0310
0.69491.41750.7051-3.3085-3.28550.4000-0.0230-51.4100-51.6195-0.7593-0.7593
0.71.62000.7007-3.3122-3.29810.4400-0.0141-51.5352-51.6561-0.7261-0.7261
0.70041.82250.7092-3.5268-3.50140.4600-0.0254-53.5688-53.8022-1.0639-1.0640
0.70562.02500.7048-3.3574-3.33770.4800-0.0197-51.9312-52.1080-0.8329-0.8329
0.68292.22750.6964-3.4182-3.41520.5400-0.0030-52.7066-52.7166-1.0186-1.0187
0.71012.43000.6992-4.3808-4.38040.5400-0.0003-62.3591-62.3421-1.3638-1.3638
0.71072.63250.7081-4.1483-4.12660.4600-0.0217-59.8212-60.0177-1.3589-1.3589
0.70352.83500.6913-3.0909-3.09660.29000.0058-49.5212-49.4432-0.7017-0.7017
0.71123.03750.7096-4.4207-4.39390.4600-0.0268-62.4938-62.7416-1.3752-1.3752
0.6593.24000.7992-4.2280-4.12900.5200-0.0990-59.8449-60.8146-1.0809-1.0815
0.62533.44250.9164-4.3837-4.11240.5200-0.2713-59.6787-62.3715-0.7324-0.7317
0.9563.64500.5266-3.8419-5.45700.68001.6151-73.1246-56.9532-0.3747-0.3742
0.56043.84750.6506-3.5933-6.21680.70002.6234-80.7223-54.4675-0.1960-0.1952
0.87764.05000.5657-3.9281-7.05640.84003.1284-89.1191-57.8147-0.6674-0.6680
0.49784.25250.6285-4.8602-10.35180.88005.4916-122.0728-67.1361-0.9244-0.9236
1.02584.45500.6966-5.0528-8.78950.80003.7367-106.4495-69.0625-0.6216-0.6205
0.35594.65750.6527-5.5366-9.70920.81004.1726-115.6466-73.9002-1.1615-1.1603
0.22364.86000.3743-5.2783-10.88810.91005.6099-127.4360-71.3169-1.0731-1.0714
0.09955.06250.1816-4.6140-10.25040.95005.6364-121.0588-64.6745-1.0550-1.0504
0.49545.26500.2771-4.9474-10.62560.90005.6781-124.8103-68.0087-0.9020-0.9007
0.20315.46750.2731-5.6955-12.69490.96006.9994-145.5037-75.4888-1.0406-1.0388
0.36655.67000.2912-5.5615-11.94340.93006.3819-137.9883-74.1489-0.9311-0.9288
0.1325.87250.2410-6.2707-13.33870.94007.0680-151.9420-81.2413-1.0742-1.0720
0.10446.07500.2450-6.0942-13.23970.95007.1455-150.9520-79.4765-1.0715-1.0693
0.19846.27750.2646-6.1961-13.47180.95007.2757-153.2727-80.4953-1.0771-1.0748
0.01566.48000.3140-6.1100-13.63770.95007.5277-154.9315-79.6341-1.1101-1.1082
0.26826.68250.2528-5.9327-13.52680.96007.5942-153.8231-77.8608-1.0893-1.0873
0.00116.88500.2762-5.9315-13.54610.95007.6146-154.0158-77.8491-1.0916-1.0895
0.10317.08750.2613-5.8587-13.53050.95007.6718-153.8600-77.1214-1.0933-1.0913
0.00347.29000.2675-5.8590-13.54900.95007.6900-154.0449-77.1244-1.0975-1.0955
0.13147.49250.2662-5.8482-13.55200.95007.7038-154.0743-77.0162-1.0978-1.0958
0.33187.69500.2651-5.8403-13.54640.95007.7060-154.0184-76.9377-1.0974-1.0954
0.10937.89750.2653-5.8449-13.54880.95007.7039-154.0427-76.9835-1.0977-1.0957
0.18088.010000.2656-5.8394-13.54640.95007.7070-154.0191-76.9285-1.0971-1.0952

Framework versions

  • —Transformers 4.40.2
  • —Pytorch 2.0.0+cu117
  • —Datasets 2.19.1
  • —Tokenizers 0.19.1