Model save

README.md

@@ -0,0 +1,79 @@
+---
+license: mit
+library_name: peft
+tags:
+- trl
+- sft
+- generated_from_trainer
+base_model: microsoft/Phi-3-mini-128k-instruct
+datasets:
+- generator
+metrics:
+- bleu
+- rouge
+model-index:
+- name: Phi-3-mini-128k-instruct-advisegpt-v0.2
+  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. -->
+
+# Phi-3-mini-128k-instruct-advisegpt-v0.2
+
+This model is a fine-tuned version of [microsoft/Phi-3-mini-128k-instruct](https://huggingface.co/microsoft/Phi-3-mini-128k-instruct) on the generator dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.8935
+- Bleu: {'bleu': 0.26234942453828036, 'precisions': [0.6386386439809577, 0.32210746013057906, 0.19439435894133555, 0.13267612303321208], 'brevity_penalty': 0.9720688221242278, 'length_ratio': 0.9724517334440523, 'translation_length': 187372, 'reference_length': 192680}
+- Rouge: {'rouge1': 0.6264335677482978, 'rouge2': 0.303034334791063, 'rougeL': 0.5025911195619426, 'rougeLsum': 0.5017835431871924}
+- Exact Match: {'exact_match': 0.0}
+
+## 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: 2e-05
+- train_batch_size: 5
+- eval_batch_size: 4
+- seed: 42
+- gradient_accumulation_steps: 12
+- total_train_batch_size: 60
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- num_epochs: 8
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss | Bleu                                                                                                                                                                                                                                                                  | Rouge                                                                                                                          | Exact Match          |
+|:-------------:|:------:|:----:|:---------------:|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------:|:------------------------------------------------------------------------------------------------------------------------------:|:--------------------:|
+| 1.0263        | 0.9930 | 71   | 1.8935          | {'bleu': 0.26234942453828036, 'precisions': [0.6386386439809577, 0.32210746013057906, 0.19439435894133555, 0.13267612303321208], 'brevity_penalty': 0.9720688221242278, 'length_ratio': 0.9724517334440523, 'translation_length': 187372, 'reference_length': 192680} | {'rouge1': 0.6264335677482978, 'rouge2': 0.303034334791063, 'rougeL': 0.5025911195619426, 'rougeLsum': 0.5017835431871924}     | {'exact_match': 0.0} |
+| 0.7406        | 2.0    | 143  | 2.0190          | {'bleu': 0.2316346526053078, 'precisions': [0.6194236274162941, 0.28977498736790047, 0.16667026013087397, 0.10975622939300578], 'brevity_penalty': 0.9676527647784504, 'length_ratio': 0.9681648328835375, 'translation_length': 186546, 'reference_length': 192680}  | {'rouge1': 0.6036218396315156, 'rouge2': 0.2682122181745471, 'rougeL': 0.47708940409367784, 'rougeLsum': 0.4770613490668666}   | {'exact_match': 0.0} |
+| 0.5882        | 2.9930 | 214  | 2.0681          | {'bleu': 0.22838404823316763, 'precisions': [0.6165055539838692, 0.2855842981089862, 0.1628873061791873, 0.10631461677977687], 'brevity_penalty': 0.9719140991086993, 'length_ratio': 0.9723012248287316, 'translation_length': 187343, 'reference_length': 192680}   | {'rouge1': 0.6006461234391669, 'rouge2': 0.2637867501761157, 'rougeL': 0.4734228347835384, 'rougeLsum': 0.4732165944934509}    | {'exact_match': 0.0} |
+| 0.5344        | 4.0    | 286  | 2.0990          | {'bleu': 0.23243181277634892, 'precisions': [0.6183425166820463, 0.290359158131201, 0.16708232101387482, 0.10871728128815054], 'brevity_penalty': 0.9726288315430208, 'length_ratio': 0.9729966784305585, 'translation_length': 187477, 'reference_length': 192680}   | {'rouge1': 0.6020165553663895, 'rouge2': 0.2689980360965313, 'rougeL': 0.4761211517574821, 'rougeLsum': 0.476013109131896}     | {'exact_match': 0.0} |
+| 0.491         | 4.9930 | 357  | 2.1029          | {'bleu': 0.23217356305609613, 'precisions': [0.6182632097844334, 0.28982060887176503, 0.16651395073437608, 0.1084203343202102], 'brevity_penalty': 0.9735242125771166, 'length_ratio': 0.9738685904089682, 'translation_length': 187645, 'reference_length': 192680}  | {'rouge1': 0.602022171746183, 'rouge2': 0.2678457021558207, 'rougeL': 0.4757373660712696, 'rougeLsum': 0.4756766948490637}     | {'exact_match': 0.0} |
+| 0.4804        | 6.0    | 429  | 2.1066          | {'bleu': 0.22688560402307306, 'precisions': [0.6157982530470419, 0.2836332155892459, 0.16120717833852222, 0.10478493323661374], 'brevity_penalty': 0.9735029030458111, 'length_ratio': 0.9738478305999585, 'translation_length': 187641, 'reference_length': 192680}  | {'rouge1': 0.5997469653277846, 'rouge2': 0.2615884826579755, 'rougeL': 0.4719633878547087, 'rougeLsum': 0.4719354595076038}    | {'exact_match': 0.0} |
+| 0.4667        | 6.9930 | 500  | 2.1083          | {'bleu': 0.2278015535859749, 'precisions': [0.6163871882176788, 0.28446401188294446, 0.1620224273628829, 0.10547183495849786], 'brevity_penalty': 0.9736627137558076, 'length_ratio': 0.9740035291675316, 'translation_length': 187671, 'reference_length': 192680}   | {'rouge1': 0.6005116792432119, 'rouge2': 0.26230752315350514, 'rougeL': 0.47195529453152857, 'rougeLsum': 0.47187802968758125} | {'exact_match': 0.0} |
+| 0.4827        | 7.9441 | 568  | 2.1093          | {'bleu': 0.2279193811355966, 'precisions': [0.6163472757204277, 0.2845680034617419, 0.16225351810881897, 0.10543872371283539], 'brevity_penalty': 0.9738224996031655, 'length_ratio': 0.9741592277351049, 'translation_length': 187701, 'reference_length': 192680}   | {'rouge1': 0.6005721411221121, 'rouge2': 0.2625293432287747, 'rougeL': 0.4722072250908843, 'rougeLsum': 0.472187239051013}     | {'exact_match': 0.0} |
+
+
+### Framework versions
+
+- PEFT 0.10.0
+- Transformers 4.40.1
+- Pytorch 2.3.0+cu121
+- Datasets 2.19.0
+- Tokenizers 0.19.1

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4cd995be8ebfad324bc30ff3d5ca7d7df1727db626f2281759ad40098a409ea5
+oid sha256:b51daf2593a428d8069acda964296c2272e473251c5000e24e3c75dbf0626a77
 size 545621264

results.json

@@ -0,0 +1,4 @@
+Pre-training results: 
+{"eval_loss": 4.0594482421875, "eval_bleu": {"bleu": 0.1826898245158488, "precisions": [0.5573254531286369, 0.22778189271183172, 0.119520655944497, 0.07341526849915855], "brevity_penalty": 1.0, "length_ratio": 1.0256695038405648, "translation_length": 197626, "reference_length": 192680}, "eval_rouge": {"rouge1": 0.5620978339462965, "rouge2": 0.21928124564678209, "rougeL": 0.4200989137725146, "rougeLsum": 0.4164644643467429}, "eval_exact_match": {"exact_match": 0.0}, "eval_runtime": 95.4065, "eval_samples_per_second": 5.367, "eval_steps_per_second": 1.342}
+ Post-training results: 
+{"eval_loss": 1.8935281038284302, "eval_bleu": {"bleu": 0.26234942453828036, "precisions": [0.6386386439809577, 0.32210746013057906, 0.19439435894133555, 0.13267612303321208], "brevity_penalty": 0.9720688221242278, "length_ratio": 0.9724517334440523, "translation_length": 187372, "reference_length": 192680}, "eval_rouge": {"rouge1": 0.6264335677482978, "rouge2": 0.303034334791063, "rougeL": 0.5025911195619426, "rougeLsum": 0.5017835431871924}, "eval_exact_match": {"exact_match": 0.0}, "eval_runtime": 94.9049, "eval_samples_per_second": 5.395, "eval_steps_per_second": 1.349, "epoch": 7.944055944055944}