File size: 4,429 Bytes
70d6617 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 |
---
base_model: microsoft/Phi-3-mini-4k-instruct
library_name: peft
license: mit
tags:
- axolotl
- generated_from_trainer
model-index:
- name: phi3-nosys-gpt4ominiplans-27k-512rank-long
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. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
# model and tokenizer
base_model: microsoft/Phi-3-mini-4k-instruct # change for model
trust_remote_code: true
sequence_len: 2048
strict: false
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer
bf16: auto
pad_to_sequence_len: true
save_safetensors: true
datasets:
- path: verifiers-for-code/sampled_10k_from_27k
type: completion
field: text_nosys_phi
train_on_split: train
val_set_size: 0.05
# lora
adapter: lora
lora_r: 512
lora_alpha: 32
lora_dropout: 0.05
lora_target_linear: true
lora_modules_to_save:
- embed_tokens
- lm_head
use_rslora: true
# logging
wandb_project: valeris
wandb_name: phi3-nosys-gpt4ominiplans-27k-512rank-long
output_dir: ./outputs/phi3-nosys-gpt4ominiplans-27k-512rank-long
gradient_accumulation_steps: 2
gradient_checkpointing: true
gradient_checkpointing_kwargs:
use_reentrant: true
micro_batch_size: 2
num_epochs: 3
eval_batch_size: 2
warmup_ratio: 0.05
learning_rate: 1e-5
lr_scheduler: cosine
optimizer: adamw_torch
hub_model_id: verifiers-for-code/phi3-nosys-gpt4ominiplans-27k-512rank-long
push_to_hub: true
hub_strategy: all_checkpoints
hub_always_push: true
evals_per_epoch: 8
saves_per_epoch: 4
logging_steps: 1
# eval_table_size: 10
# eval_max_new_tokens: 512
tokens: ["<thinking>", "</thinking>", "<plan>", "</plan>"]
special_tokens:
pad_token: "<|endoftext|>"
```
</details><br>
# phi3-nosys-gpt4ominiplans-27k-512rank-long
This model is a fine-tuned version of [microsoft/Phi-3-mini-4k-instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6378
## 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
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- total_eval_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 44
- num_epochs: 3
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 1.0833 | 0.0034 | 1 | 1.0330 |
| 1.0093 | 0.1279 | 38 | 0.9910 |
| 0.9169 | 0.2559 | 76 | 0.8668 |
| 0.795 | 0.3838 | 114 | 0.7676 |
| 0.6999 | 0.5118 | 152 | 0.7243 |
| 0.7246 | 0.6397 | 190 | 0.6989 |
| 0.6873 | 0.7677 | 228 | 0.6816 |
| 0.7014 | 0.8956 | 266 | 0.6687 |
| 0.6586 | 1.0236 | 304 | 0.6585 |
| 0.6532 | 1.1515 | 342 | 0.6511 |
| 0.6334 | 1.2795 | 380 | 0.6463 |
| 0.5968 | 1.4074 | 418 | 0.6434 |
| 0.6366 | 1.5354 | 456 | 0.6414 |
| 0.6126 | 1.6633 | 494 | 0.6400 |
| 0.6564 | 1.7912 | 532 | 0.6391 |
| 0.6296 | 1.9192 | 570 | 0.6387 |
| 0.6225 | 2.0471 | 608 | 0.6383 |
| 0.6354 | 2.1751 | 646 | 0.6381 |
| 0.6111 | 2.3030 | 684 | 0.6379 |
| 0.5899 | 2.4310 | 722 | 0.6378 |
| 0.6415 | 2.5589 | 760 | 0.6378 |
| 0.6443 | 2.6869 | 798 | 0.6377 |
| 0.6103 | 2.8148 | 836 | 0.6377 |
| 0.6451 | 2.9428 | 874 | 0.6378 |
### Framework versions
- PEFT 0.11.1
- Transformers 4.44.0.dev0
- Pytorch 2.4.0
- Datasets 2.19.1
- Tokenizers 0.19.1 |