Update README.md
Browse files
README.md
CHANGED
@@ -11,17 +11,14 @@ should probably proofread and complete it, then remove this comment. -->
|
|
11 |
|
12 |
# gpt-fake-lang-17m
|
13 |
|
14 |
-
This model is a pre-trained
|
|
|
15 |
It achieves the following results on the evaluation set:
|
16 |
- Loss: 3.5592
|
17 |
|
18 |
-
## Model description
|
19 |
-
|
20 |
-
More information needed
|
21 |
-
|
22 |
## Intended uses & limitations
|
23 |
|
24 |
-
|
25 |
|
26 |
## Training and evaluation data
|
27 |
|
@@ -33,12 +30,9 @@ More information needed
|
|
33 |
|
34 |
The following hyperparameters were used during training:
|
35 |
- learning_rate: 0.001
|
36 |
-
-
|
37 |
-
- eval_batch_size: 4
|
38 |
- seed: 42
|
39 |
-
-
|
40 |
-
- total_train_batch_size: 64
|
41 |
-
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
42 |
- lr_scheduler_type: linear
|
43 |
- num_epochs: 1
|
44 |
|
|
|
11 |
|
12 |
# gpt-fake-lang-17m
|
13 |
|
14 |
+
This model is a pre-trained GPTJ (with 17m parameters) on a synthetic dataset (1gb of documents created in 4 fake languages, each with a formal and informal writing style) for 1 epoch.
|
15 |
+
|
16 |
It achieves the following results on the evaluation set:
|
17 |
- Loss: 3.5592
|
18 |
|
|
|
|
|
|
|
|
|
19 |
## Intended uses & limitations
|
20 |
|
21 |
+
This model is to be used as a base model for fine-tuning any language/task to probe the effectiveness of both pre-training on an algorithmically generated corpus and extremely small models. It can only generate text based on its training data (which will be uploaded as a huggingface dataset soon).
|
22 |
|
23 |
## Training and evaluation data
|
24 |
|
|
|
30 |
|
31 |
The following hyperparameters were used during training:
|
32 |
- learning_rate: 0.001
|
33 |
+
- batch_size 64
|
|
|
34 |
- seed: 42
|
35 |
+
- optimizer: Adam
|
|
|
|
|
36 |
- lr_scheduler_type: linear
|
37 |
- num_epochs: 1
|
38 |
|