Model save

README.md

@@ -0,0 +1,77 @@
+---
+license: apache-2.0
+library_name: peft
+tags:
+- trl
+- dpo
+- llama-factory
+- generated_from_trainer
+base_model: tiiuae/falcon-7b-instruct
+model-index:
+- name: Falcon-7B-Instruct-ORPO-SALT
+  results: []
+---
+
+<!-- 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. -->
+
+# Falcon-7B-Instruct-ORPO-SALT
+
+This model is a fine-tuned version of [tiiuae/falcon-7b-instruct](https://huggingface.co/tiiuae/falcon-7b-instruct) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.4485
+- Rewards/chosen: -0.1373
+- Rewards/rejected: -0.1429
+- Rewards/accuracies: 0.4809
+- Rewards/margins: 0.0056
+- Logps/rejected: -1.4290
+- Logps/chosen: -1.3726
+- Logits/rejected: -14.3178
+- Logits/chosen: -14.2778
+- Sft Loss: 1.3726
+- Odds Ratio Loss: 0.7590
+
+## 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: 5e-06
+- train_batch_size: 2
+- eval_batch_size: 2
+- seed: 42
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 16
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_steps: 0.1
+- num_epochs: 3.0
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen | Sft Loss | Odds Ratio Loss |
+|:-------------:|:------:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|:--------:|:---------------:|
+| 1.5005        | 0.8082 | 500  | 1.5202          | -0.1444        | -0.1490          | 0.4818             | 0.0046          | -1.4898        | -1.4436      | -14.2657        | -14.2276      | 1.4436   | 0.7658          |
+| 1.3401        | 1.6165 | 1000 | 1.4635          | -0.1387        | -0.1442          | 0.4836             | 0.0055          | -1.4423        | -1.3875      | -14.3083        | -14.2685      | 1.3875   | 0.7603          |
+| 1.446         | 2.4247 | 1500 | 1.4485          | -0.1373        | -0.1429          | 0.4809             | 0.0056          | -1.4290        | -1.3726      | -14.3178        | -14.2778      | 1.3726   | 0.7590          |
+
+
+### Framework versions
+
+- PEFT 0.10.0
+- Transformers 4.40.1
+- Pytorch 2.3.0
+- Datasets 2.19.0
+- Tokenizers 0.19.1

trainer_log.jsonl

@@ -151,3 +151,39 @@
 {"current_steps": 1490, "total_steps": 1854, "loss": 1.5016, "accuracy": 0.4749999940395355, "learning_rate": 4.607082849092523e-07, "epoch": 2.40856738735098, "percentage": 80.37, "elapsed_time": "4:13:34", "remaining_time": "1:01:56"}
 {"current_steps": 1500, "total_steps": 1854, "loss": 1.446, "accuracy": 0.44999998807907104, "learning_rate": 4.3649635614901405e-07, "epoch": 2.4247322691452817, "percentage": 80.91, "elapsed_time": "4:16:33", "remaining_time": "1:00:32"}
 {"current_steps": 1500, "total_steps": 1854, "eval_loss": 1.4484930038452148, "epoch": 2.4247322691452817, "percentage": 80.91, "elapsed_time": "4:23:10", "remaining_time": "1:02:06"}
+{"current_steps": 1510, "total_steps": 1854, "loss": 1.4053, "accuracy": 0.53125, "learning_rate": 4.128769732701973e-07, "epoch": 2.4408971509395836, "percentage": 81.45, "elapsed_time": "4:25:56", "remaining_time": "1:00:35"}
+{"current_steps": 1520, "total_steps": 1854, "loss": 1.3737, "accuracy": 0.48124998807907104, "learning_rate": 3.8985691870233046e-07, "epoch": 2.4570620327338855, "percentage": 81.98, "elapsed_time": "4:28:49", "remaining_time": "0:59:04"}
+{"current_steps": 1530, "total_steps": 1854, "loss": 1.5046, "accuracy": 0.45625001192092896, "learning_rate": 3.6744280277467904e-07, "epoch": 2.4732269145281873, "percentage": 82.52, "elapsed_time": "4:31:40", "remaining_time": "0:57:31"}
+{"current_steps": 1540, "total_steps": 1854, "loss": 1.2927, "accuracy": 0.543749988079071, "learning_rate": 3.456410618180503e-07, "epoch": 2.489391796322489, "percentage": 83.06, "elapsed_time": "4:34:37", "remaining_time": "0:55:59"}
+{"current_steps": 1550, "total_steps": 1854, "loss": 1.3673, "accuracy": 0.518750011920929, "learning_rate": 3.244579563165753e-07, "epoch": 2.5055566781167915, "percentage": 83.6, "elapsed_time": "4:37:32", "remaining_time": "0:54:26"}
+{"current_steps": 1560, "total_steps": 1854, "loss": 1.4344, "accuracy": 0.4749999940395355, "learning_rate": 3.038995691099697e-07, "epoch": 2.521721559911093, "percentage": 84.14, "elapsed_time": "4:40:23", "remaining_time": "0:52:50"}
+{"current_steps": 1570, "total_steps": 1854, "loss": 1.5307, "accuracy": 0.5, "learning_rate": 2.839718036468192e-07, "epoch": 2.5378864417053952, "percentage": 84.68, "elapsed_time": "4:43:23", "remaining_time": "0:51:15"}
+{"current_steps": 1580, "total_steps": 1854, "loss": 1.5325, "accuracy": 0.5, "learning_rate": 2.646803822893723e-07, "epoch": 2.5540513234996967, "percentage": 85.22, "elapsed_time": "4:46:23", "remaining_time": "0:49:39"}
+{"current_steps": 1590, "total_steps": 1854, "loss": 1.4097, "accuracy": 0.518750011920929, "learning_rate": 2.460308446703341e-07, "epoch": 2.570216205293999, "percentage": 85.76, "elapsed_time": "4:49:18", "remaining_time": "0:48:02"}
+{"current_steps": 1600, "total_steps": 1854, "loss": 1.3864, "accuracy": 0.5062500238418579, "learning_rate": 2.2802854610213143e-07, "epoch": 2.5863810870883004, "percentage": 86.3, "elapsed_time": "4:52:13", "remaining_time": "0:46:23"}
+{"current_steps": 1610, "total_steps": 1854, "loss": 1.4698, "accuracy": 0.4937500059604645, "learning_rate": 2.106786560391072e-07, "epoch": 2.6025459688826027, "percentage": 86.84, "elapsed_time": "4:55:04", "remaining_time": "0:44:43"}
+{"current_steps": 1620, "total_steps": 1854, "loss": 1.3982, "accuracy": 0.5375000238418579, "learning_rate": 1.9398615659308255e-07, "epoch": 2.6187108506769046, "percentage": 87.38, "elapsed_time": "4:58:05", "remaining_time": "0:43:03"}
+{"current_steps": 1630, "total_steps": 1854, "loss": 1.4475, "accuracy": 0.46875, "learning_rate": 1.7795584110272184e-07, "epoch": 2.6348757324712064, "percentage": 87.92, "elapsed_time": "5:01:17", "remaining_time": "0:41:24"}
+{"current_steps": 1640, "total_steps": 1854, "loss": 1.4028, "accuracy": 0.4937500059604645, "learning_rate": 1.6259231275709636e-07, "epoch": 2.6510406142655083, "percentage": 88.46, "elapsed_time": "5:04:19", "remaining_time": "0:39:42"}
+{"current_steps": 1650, "total_steps": 1854, "loss": 1.368, "accuracy": 0.5375000238418579, "learning_rate": 1.478999832738548e-07, "epoch": 2.66720549605981, "percentage": 89.0, "elapsed_time": "5:07:12", "remaining_time": "0:37:58"}
+{"current_steps": 1660, "total_steps": 1854, "loss": 1.383, "accuracy": 0.48750001192092896, "learning_rate": 1.338830716323769e-07, "epoch": 2.683370377854112, "percentage": 89.54, "elapsed_time": "5:10:20", "remaining_time": "0:36:16"}
+{"current_steps": 1670, "total_steps": 1854, "loss": 1.3249, "accuracy": 0.512499988079071, "learning_rate": 1.205456028622723e-07, "epoch": 2.699535259648414, "percentage": 90.08, "elapsed_time": "5:13:10", "remaining_time": "0:34:30"}
+{"current_steps": 1680, "total_steps": 1854, "loss": 1.3983, "accuracy": 0.5874999761581421, "learning_rate": 1.0789140688756805e-07, "epoch": 2.7157001414427158, "percentage": 90.61, "elapsed_time": "5:16:08", "remaining_time": "0:32:44"}
+{"current_steps": 1690, "total_steps": 1854, "loss": 1.3598, "accuracy": 0.46875, "learning_rate": 9.592411742693098e-08, "epoch": 2.7318650232370176, "percentage": 91.15, "elapsed_time": "5:19:00", "remaining_time": "0:30:57"}
+{"current_steps": 1700, "total_steps": 1854, "loss": 1.3626, "accuracy": 0.574999988079071, "learning_rate": 8.464717095022168e-08, "epoch": 2.7480299050313195, "percentage": 91.69, "elapsed_time": "5:21:57", "remaining_time": "0:29:09"}
+{"current_steps": 1710, "total_steps": 1854, "loss": 1.4574, "accuracy": 0.46875, "learning_rate": 7.406380569169841e-08, "epoch": 2.7641947868256214, "percentage": 92.23, "elapsed_time": "5:24:56", "remaining_time": "0:27:21"}
+{"current_steps": 1720, "total_steps": 1854, "loss": 1.4887, "accuracy": 0.48750001192092896, "learning_rate": 6.417706072013808e-08, "epoch": 2.7803596686199232, "percentage": 92.77, "elapsed_time": "5:27:57", "remaining_time": "0:25:33"}
+{"current_steps": 1730, "total_steps": 1854, "loss": 1.4818, "accuracy": 0.41874998807907104, "learning_rate": 5.498977506615294e-08, "epoch": 2.796524550414225, "percentage": 93.31, "elapsed_time": "5:31:06", "remaining_time": "0:23:43"}
+{"current_steps": 1740, "total_steps": 1854, "loss": 1.4204, "accuracy": 0.5687500238418579, "learning_rate": 4.6504586906947756e-08, "epoch": 2.812689432208527, "percentage": 93.85, "elapsed_time": "5:34:09", "remaining_time": "0:21:53"}
+{"current_steps": 1750, "total_steps": 1854, "loss": 1.4888, "accuracy": 0.4375, "learning_rate": 3.8723932808754914e-08, "epoch": 2.828854314002829, "percentage": 94.39, "elapsed_time": "5:37:16", "remaining_time": "0:20:02"}
+{"current_steps": 1760, "total_steps": 1854, "loss": 1.3941, "accuracy": 0.5, "learning_rate": 3.1650047027158014e-08, "epoch": 2.8450191957971307, "percentage": 94.93, "elapsed_time": "5:40:25", "remaining_time": "0:18:10"}
+{"current_steps": 1770, "total_steps": 1854, "loss": 1.3373, "accuracy": 0.48750001192092896, "learning_rate": 2.5284960865517848e-08, "epoch": 2.8611840775914326, "percentage": 95.47, "elapsed_time": "5:43:36", "remaining_time": "0:16:18"}
+{"current_steps": 1780, "total_steps": 1854, "loss": 1.4034, "accuracy": 0.581250011920929, "learning_rate": 1.9630502091670388e-08, "epoch": 2.8773489593857344, "percentage": 96.01, "elapsed_time": "5:46:31", "remaining_time": "0:14:24"}
+{"current_steps": 1790, "total_steps": 1854, "loss": 1.2918, "accuracy": 0.5, "learning_rate": 1.4688294413074677e-08, "epoch": 2.8935138411800363, "percentage": 96.55, "elapsed_time": "5:49:26", "remaining_time": "0:12:29"}
+{"current_steps": 1800, "total_steps": 1854, "loss": 1.3902, "accuracy": 0.5062500238418579, "learning_rate": 1.0459757010556626e-08, "epoch": 2.909678722974338, "percentage": 97.09, "elapsed_time": "5:52:24", "remaining_time": "0:10:34"}
+{"current_steps": 1810, "total_steps": 1854, "loss": 1.4669, "accuracy": 0.4937500059604645, "learning_rate": 6.94610413078306e-09, "epoch": 2.92584360476864, "percentage": 97.63, "elapsed_time": "5:55:18", "remaining_time": "0:08:38"}
+{"current_steps": 1820, "total_steps": 1854, "loss": 1.3162, "accuracy": 0.5625, "learning_rate": 4.14834473758563e-09, "epoch": 2.942008486562942, "percentage": 98.17, "elapsed_time": "5:58:21", "remaining_time": "0:06:41"}
+{"current_steps": 1830, "total_steps": 1854, "loss": 1.3957, "accuracy": 0.518750011920929, "learning_rate": 2.067282222230349e-09, "epoch": 2.9581733683572438, "percentage": 98.71, "elapsed_time": "6:01:37", "remaining_time": "0:04:44"}
+{"current_steps": 1840, "total_steps": 1854, "loss": 1.3287, "accuracy": 0.550000011920929, "learning_rate": 7.035141727212979e-10, "epoch": 2.9743382501515456, "percentage": 99.24, "elapsed_time": "6:04:24", "remaining_time": "0:02:46"}
+{"current_steps": 1850, "total_steps": 1854, "loss": 1.4375, "accuracy": 0.4937500059604645, "learning_rate": 5.743220219761592e-11, "epoch": 2.9905031319458475, "percentage": 99.78, "elapsed_time": "6:07:29", "remaining_time": "0:00:47"}
+{"current_steps": 1854, "total_steps": 1854, "epoch": 2.9969690846635686, "percentage": 100.0, "elapsed_time": "6:08:42", "remaining_time": "0:00:00"}