robinsmits
commited on
Commit
•
5eb7542
1
Parent(s):
fc52456
update model card README.md
Browse files
README.md
CHANGED
@@ -1,22 +1,11 @@
|
|
1 |
---
|
2 |
-
license:
|
3 |
-
inference: false
|
4 |
-
datasets:
|
5 |
-
- BramVanroy/alpaca-cleaned-dutch
|
6 |
base_model: DAMO-NLP-MT/polylm-1.7b
|
7 |
tags:
|
8 |
- generated_from_trainer
|
9 |
-
- alpaca
|
10 |
-
- Transformers
|
11 |
-
- PolyLM
|
12 |
-
- text-generation-inference
|
13 |
model-index:
|
14 |
- name: polylm_1.7b_ft_alpaca_clean_dutch
|
15 |
results: []
|
16 |
-
language:
|
17 |
-
- nl
|
18 |
-
library_name: peft
|
19 |
-
pipeline_tag: text-generation
|
20 |
---
|
21 |
|
22 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -26,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
26 |
|
27 |
This model is a fine-tuned version of [DAMO-NLP-MT/polylm-1.7b](https://huggingface.co/DAMO-NLP-MT/polylm-1.7b) on an unknown dataset.
|
28 |
It achieves the following results on the evaluation set:
|
29 |
-
- Loss: 1.
|
30 |
|
31 |
## Model description
|
32 |
|
@@ -42,21 +31,10 @@ More information needed
|
|
42 |
|
43 |
## Training procedure
|
44 |
|
45 |
-
|
46 |
-
The following `bitsandbytes` quantization config was used during training:
|
47 |
-
- load_in_8bit: False
|
48 |
-
- load_in_4bit: True
|
49 |
-
- llm_int8_threshold: 6.0
|
50 |
-
- llm_int8_skip_modules: None
|
51 |
-
- llm_int8_enable_fp32_cpu_offload: False
|
52 |
-
- llm_int8_has_fp16_weight: False
|
53 |
-
- bnb_4bit_quant_type: nf4
|
54 |
-
- bnb_4bit_use_double_quant: True
|
55 |
-
- bnb_4bit_compute_dtype: bfloat16
|
56 |
### Training hyperparameters
|
57 |
|
58 |
The following hyperparameters were used during training:
|
59 |
-
- learning_rate: 0.
|
60 |
- train_batch_size: 8
|
61 |
- eval_batch_size: 8
|
62 |
- seed: 42
|
@@ -71,25 +49,23 @@ The following hyperparameters were used during training:
|
|
71 |
|
72 |
| Training Loss | Epoch | Step | Validation Loss |
|
73 |
|:-------------:|:-----:|:----:|:---------------:|
|
74 |
-
| 2.
|
75 |
-
| 2.
|
76 |
-
|
|
77 |
-
| 1.
|
78 |
-
| 1.
|
79 |
-
| 1.
|
80 |
-
| 1.
|
81 |
-
| 1.
|
82 |
-
| 1.
|
83 |
-
| 1.
|
84 |
-
| 1.
|
85 |
-
| 1.
|
86 |
|
87 |
|
88 |
### Framework versions
|
89 |
|
90 |
-
- PEFT 0.4.0
|
91 |
- Transformers 4.31.0
|
92 |
- Pytorch 2.0.1+cu118
|
93 |
- Datasets 2.13.1
|
94 |
- Tokenizers 0.13.3
|
95 |
-
- PEFT 0.4.0
|
|
|
1 |
---
|
2 |
+
license: apache-2.0
|
|
|
|
|
|
|
3 |
base_model: DAMO-NLP-MT/polylm-1.7b
|
4 |
tags:
|
5 |
- generated_from_trainer
|
|
|
|
|
|
|
|
|
6 |
model-index:
|
7 |
- name: polylm_1.7b_ft_alpaca_clean_dutch
|
8 |
results: []
|
|
|
|
|
|
|
|
|
9 |
---
|
10 |
|
11 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
15 |
|
16 |
This model is a fine-tuned version of [DAMO-NLP-MT/polylm-1.7b](https://huggingface.co/DAMO-NLP-MT/polylm-1.7b) on an unknown dataset.
|
17 |
It achieves the following results on the evaluation set:
|
18 |
+
- Loss: 1.8483
|
19 |
|
20 |
## Model description
|
21 |
|
|
|
31 |
|
32 |
## Training procedure
|
33 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
34 |
### Training hyperparameters
|
35 |
|
36 |
The following hyperparameters were used during training:
|
37 |
+
- learning_rate: 0.0001
|
38 |
- train_batch_size: 8
|
39 |
- eval_batch_size: 8
|
40 |
- seed: 42
|
|
|
49 |
|
50 |
| Training Loss | Epoch | Step | Validation Loss |
|
51 |
|:-------------:|:-----:|:----:|:---------------:|
|
52 |
+
| 2.1248 | 0.16 | 128 | 2.1129 |
|
53 |
+
| 2.0512 | 0.33 | 256 | 2.0347 |
|
54 |
+
| 1.9983 | 0.49 | 384 | 1.9948 |
|
55 |
+
| 1.9557 | 0.66 | 512 | 1.9655 |
|
56 |
+
| 1.9583 | 0.82 | 640 | 1.9386 |
|
57 |
+
| 1.916 | 0.99 | 768 | 1.9177 |
|
58 |
+
| 1.8671 | 1.15 | 896 | 1.9019 |
|
59 |
+
| 1.8626 | 1.32 | 1024 | 1.8885 |
|
60 |
+
| 1.8321 | 1.48 | 1152 | 1.8762 |
|
61 |
+
| 1.8596 | 1.65 | 1280 | 1.8631 |
|
62 |
+
| 1.843 | 1.81 | 1408 | 1.8539 |
|
63 |
+
| 1.8333 | 1.98 | 1536 | 1.8483 |
|
64 |
|
65 |
|
66 |
### Framework versions
|
67 |
|
|
|
68 |
- Transformers 4.31.0
|
69 |
- Pytorch 2.0.1+cu118
|
70 |
- Datasets 2.13.1
|
71 |
- Tokenizers 0.13.3
|
|