mzamfirdaus
commited on
Commit
•
e9a47a7
1
Parent(s):
a08481b
mzamfirdaus/test-peft-train
Browse files- README.md +101 -0
- adapter_config.json +31 -0
- adapter_model.safetensors +3 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,101 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
library_name: peft
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
base_model: microsoft/phi-2
|
7 |
+
model-index:
|
8 |
+
- name: final-checkpoint
|
9 |
+
results: []
|
10 |
+
---
|
11 |
+
|
12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
13 |
+
should probably proofread and complete it, then remove this comment. -->
|
14 |
+
|
15 |
+
# final-checkpoint
|
16 |
+
|
17 |
+
This model is a fine-tuned version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) on an unknown dataset.
|
18 |
+
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 1.3170
|
20 |
+
|
21 |
+
## Model description
|
22 |
+
|
23 |
+
More information needed
|
24 |
+
|
25 |
+
## Intended uses & limitations
|
26 |
+
|
27 |
+
More information needed
|
28 |
+
|
29 |
+
## Training and evaluation data
|
30 |
+
|
31 |
+
More information needed
|
32 |
+
|
33 |
+
## Training procedure
|
34 |
+
|
35 |
+
### Training hyperparameters
|
36 |
+
|
37 |
+
The following hyperparameters were used during training:
|
38 |
+
- learning_rate: 0.0002
|
39 |
+
- train_batch_size: 1
|
40 |
+
- eval_batch_size: 8
|
41 |
+
- seed: 42
|
42 |
+
- gradient_accumulation_steps: 4
|
43 |
+
- total_train_batch_size: 4
|
44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
45 |
+
- lr_scheduler_type: linear
|
46 |
+
- lr_scheduler_warmup_steps: 1
|
47 |
+
- training_steps: 1000
|
48 |
+
|
49 |
+
### Training results
|
50 |
+
|
51 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
52 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
53 |
+
| 1.6634 | 0.05 | 25 | 1.3938 |
|
54 |
+
| 1.1888 | 0.1 | 50 | 1.3917 |
|
55 |
+
| 1.4465 | 0.15 | 75 | 1.3522 |
|
56 |
+
| 1.205 | 0.2 | 100 | 1.3718 |
|
57 |
+
| 1.4403 | 0.25 | 125 | 1.3429 |
|
58 |
+
| 1.1372 | 0.3 | 150 | 1.3639 |
|
59 |
+
| 1.4064 | 0.35 | 175 | 1.3408 |
|
60 |
+
| 1.1466 | 0.4 | 200 | 1.3473 |
|
61 |
+
| 1.4435 | 0.45 | 225 | 1.3355 |
|
62 |
+
| 1.2273 | 0.5 | 250 | 1.3414 |
|
63 |
+
| 1.4578 | 0.55 | 275 | 1.3364 |
|
64 |
+
| 1.1637 | 0.6 | 300 | 1.3362 |
|
65 |
+
| 1.429 | 0.65 | 325 | 1.3305 |
|
66 |
+
| 1.2027 | 0.7 | 350 | 1.3306 |
|
67 |
+
| 1.3971 | 0.75 | 375 | 1.3287 |
|
68 |
+
| 1.184 | 0.8 | 400 | 1.3312 |
|
69 |
+
| 1.4383 | 0.85 | 425 | 1.3262 |
|
70 |
+
| 1.1996 | 0.9 | 450 | 1.3271 |
|
71 |
+
| 1.448 | 0.95 | 475 | 1.3250 |
|
72 |
+
| 1.1902 | 1.0 | 500 | 1.3277 |
|
73 |
+
| 1.4134 | 1.05 | 525 | 1.3237 |
|
74 |
+
| 1.1459 | 1.1 | 550 | 1.3263 |
|
75 |
+
| 1.3473 | 1.15 | 575 | 1.3225 |
|
76 |
+
| 1.1879 | 1.2 | 600 | 1.3232 |
|
77 |
+
| 1.3742 | 1.25 | 625 | 1.3217 |
|
78 |
+
| 1.1129 | 1.3 | 650 | 1.3225 |
|
79 |
+
| 1.3756 | 1.35 | 675 | 1.3201 |
|
80 |
+
| 1.1268 | 1.4 | 700 | 1.3216 |
|
81 |
+
| 1.4186 | 1.45 | 725 | 1.3200 |
|
82 |
+
| 1.1602 | 1.5 | 750 | 1.3196 |
|
83 |
+
| 1.4368 | 1.55 | 775 | 1.3189 |
|
84 |
+
| 1.1397 | 1.6 | 800 | 1.3193 |
|
85 |
+
| 1.3991 | 1.65 | 825 | 1.3186 |
|
86 |
+
| 1.1293 | 1.7 | 850 | 1.3190 |
|
87 |
+
| 1.4509 | 1.75 | 875 | 1.3177 |
|
88 |
+
| 1.1433 | 1.8 | 900 | 1.3175 |
|
89 |
+
| 1.3638 | 1.85 | 925 | 1.3170 |
|
90 |
+
| 1.144 | 1.9 | 950 | 1.3172 |
|
91 |
+
| 1.3376 | 1.95 | 975 | 1.3170 |
|
92 |
+
| 1.1152 | 2.0 | 1000 | 1.3170 |
|
93 |
+
|
94 |
+
|
95 |
+
### Framework versions
|
96 |
+
|
97 |
+
- PEFT 0.10.0
|
98 |
+
- Transformers 4.39.3
|
99 |
+
- Pytorch 2.1.2+cu121
|
100 |
+
- Datasets 2.18.0
|
101 |
+
- Tokenizers 0.15.0
|
adapter_config.json
ADDED
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "microsoft/phi-2",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layer_replication": null,
|
10 |
+
"layers_pattern": null,
|
11 |
+
"layers_to_transform": null,
|
12 |
+
"loftq_config": {},
|
13 |
+
"lora_alpha": 32,
|
14 |
+
"lora_dropout": 0.05,
|
15 |
+
"megatron_config": null,
|
16 |
+
"megatron_core": "megatron.core",
|
17 |
+
"modules_to_save": null,
|
18 |
+
"peft_type": "LORA",
|
19 |
+
"r": 32,
|
20 |
+
"rank_pattern": {},
|
21 |
+
"revision": null,
|
22 |
+
"target_modules": [
|
23 |
+
"k_proj",
|
24 |
+
"v_proj",
|
25 |
+
"q_proj",
|
26 |
+
"dense"
|
27 |
+
],
|
28 |
+
"task_type": "CAUSAL_LM",
|
29 |
+
"use_dora": false,
|
30 |
+
"use_rslora": false
|
31 |
+
}
|
adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f37e419bdc4e96d362c82b7d19d42545fea372c315898289c7c2d3cf5bf80c54
|
3 |
+
size 83920464
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:06f8d585734f31bfcae63cd51e26b61a1d05f1a94d1e01eb362c87e48a0839bb
|
3 |
+
size 4920
|