PEFT
PyTorch
mixtral
Generated from Trainer
File size: 2,823 Bytes
cdb02e6
ec7e02f
85381a2
 
 
ad4fc9c
 
 
85381a2
 
ec7e02f
85381a2
cdb02e6
85381a2
ec7e02f
8764853
de4d5cd
630b929
ec7e02f
 
85381a2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7658da5
85381a2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ad4fc9c
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
license: gpl-3.0
library_name: peft
tags:
- generated_from_trainer
datasets:
- iamshnoo/alpaca-cleaned-albanian
- noxneural/lilium_albanicum_eng_alb
base_model: nisten/shqiponja-15b-v1
model-index:
- name: shqiponja-15
  results: []
---


![image/png](https://cdn-uploads.huggingface.co/production/uploads/6379683a81c1783a4a2ddba8/V0mt5q-kb0yFeeGFNGv0q.png)


**15.6b 2expert MoE**

<!-- 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/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
<details><summary>See axolotl config</summary>

axolotl version: `0.4.0`
```yaml
base_model: nisten/shqiponja15
model_type: AutoModelForCausalLM
tokenizer_type: LlamaTokenizer
trust_remote_code: true
load_in_8bit: false
load_in_4bit: true
strict: false

datasets:
  - path: iamshnoo/alpaca-cleaned-albanian
    type: alpaca
    shards: 10
  - path: noxneural/lilium_albanicum_eng_alb
    shards: 20
    type:
      field_system: system
      field_instruction: question
      field_output: response
      format: "[INST] {instruction} [/INST]"
dataset_prepared_path: last_run_prepared
val_set_size: 0.0
output_dir: ./alora-out

#  - model.layers.2[7-9]+.block_sparse_moe.experts.*
#  - model.layers.3[0-9]+.block_sparse_moe.experts.*
#  - model.layers.2[7-9]+.b
</details><br>
```
# alora-out
## 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: 0.0002
- train_batch_size: 10
- eval_batch_size: 10
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 2
- total_train_batch_size: 80
- total_eval_batch_size: 40
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 10
- num_epochs: 3
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_nisten__shqiponja-15b-v1)

|             Metric              |Value|
|---------------------------------|----:|
|Avg.                             |71.03|
|AI2 Reasoning Challenge (25-Shot)|66.38|
|HellaSwag (10-Shot)              |85.26|
|MMLU (5-Shot)                    |64.62|
|TruthfulQA (0-shot)              |56.81|
|Winogrande (5-shot)              |84.06|
|GSM8k (5-shot)                   |69.07|