Model save
Browse files
README.md
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: unsloth/Mistral-Nemo-Base-2407-bnb-4bit
|
| 3 |
+
datasets:
|
| 4 |
+
- generator
|
| 5 |
+
library_name: peft
|
| 6 |
+
license: apache-2.0
|
| 7 |
+
tags:
|
| 8 |
+
- trl
|
| 9 |
+
- sft
|
| 10 |
+
- unsloth
|
| 11 |
+
- generated_from_trainer
|
| 12 |
+
model-index:
|
| 13 |
+
- name: mistrial_nemo_output
|
| 14 |
+
results: []
|
| 15 |
+
---
|
| 16 |
+
|
| 17 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 18 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 19 |
+
|
| 20 |
+
# mistrial_nemo_output
|
| 21 |
+
|
| 22 |
+
This model is a fine-tuned version of [unsloth/Mistral-Nemo-Base-2407-bnb-4bit](https://huggingface.co/unsloth/Mistral-Nemo-Base-2407-bnb-4bit) on the generator dataset.
|
| 23 |
+
It achieves the following results on the evaluation set:
|
| 24 |
+
- Loss: 1.5118
|
| 25 |
+
|
| 26 |
+
## Model description
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Intended uses & limitations
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training and evaluation data
|
| 35 |
+
|
| 36 |
+
More information needed
|
| 37 |
+
|
| 38 |
+
## Training procedure
|
| 39 |
+
|
| 40 |
+
### Training hyperparameters
|
| 41 |
+
|
| 42 |
+
The following hyperparameters were used during training:
|
| 43 |
+
- learning_rate: 0.0002
|
| 44 |
+
- train_batch_size: 4
|
| 45 |
+
- eval_batch_size: 4
|
| 46 |
+
- seed: 100
|
| 47 |
+
- distributed_type: multi-GPU
|
| 48 |
+
- num_devices: 2
|
| 49 |
+
- gradient_accumulation_steps: 8
|
| 50 |
+
- total_train_batch_size: 64
|
| 51 |
+
- total_eval_batch_size: 8
|
| 52 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 53 |
+
- lr_scheduler_type: cosine
|
| 54 |
+
- num_epochs: 1.0
|
| 55 |
+
|
| 56 |
+
### Training results
|
| 57 |
+
|
| 58 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 60 |
+
| 1.7287 | 0.0516 | 20 | 1.7033 |
|
| 61 |
+
| 1.6508 | 0.1033 | 40 | 1.6490 |
|
| 62 |
+
| 1.6242 | 0.1549 | 60 | 1.6253 |
|
| 63 |
+
| 1.6216 | 0.2066 | 80 | 1.6089 |
|
| 64 |
+
| 1.619 | 0.2582 | 100 | 1.5958 |
|
| 65 |
+
| 1.5579 | 0.3099 | 120 | 1.5842 |
|
| 66 |
+
| 1.5578 | 0.3615 | 140 | 1.5739 |
|
| 67 |
+
| 1.5515 | 0.4132 | 160 | 1.5641 |
|
| 68 |
+
| 1.5739 | 0.4648 | 180 | 1.5550 |
|
| 69 |
+
| 1.5669 | 0.5165 | 200 | 1.5460 |
|
| 70 |
+
| 1.5601 | 0.5681 | 220 | 1.5380 |
|
| 71 |
+
| 1.5392 | 0.6198 | 240 | 1.5310 |
|
| 72 |
+
| 1.5321 | 0.6714 | 260 | 1.5251 |
|
| 73 |
+
| 1.5326 | 0.7230 | 280 | 1.5201 |
|
| 74 |
+
| 1.5197 | 0.7747 | 300 | 1.5165 |
|
| 75 |
+
| 1.5229 | 0.8263 | 320 | 1.5142 |
|
| 76 |
+
| 1.4988 | 0.8780 | 340 | 1.5127 |
|
| 77 |
+
| 1.5044 | 0.9296 | 360 | 1.5119 |
|
| 78 |
+
| 1.5105 | 0.9813 | 380 | 1.5118 |
|
| 79 |
+
|
| 80 |
+
|
| 81 |
+
### Framework versions
|
| 82 |
+
|
| 83 |
+
- PEFT 0.12.1.dev0
|
| 84 |
+
- Transformers 4.45.0.dev0
|
| 85 |
+
- Pytorch 2.4.0+cu121
|
| 86 |
+
- Datasets 2.21.0
|
| 87 |
+
- Tokenizers 0.19.1
|
runs/Sep02_00-06-57_autodl-container-a3d5118ffa-b551dd99/events.out.tfevents.1725206856.autodl-container-a3d5118ffa-b551dd99.1032.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2cac56eca9e4e0cbeb03b876eeda8a6f438c025834fdb41fd78b8ebda76d6b02
|
| 3 |
+
size 27603
|