DeividasM
/

gpt2_lithuanian_small

@@ -7,20 +7,20 @@ license: apache-2.0
 tags:
   - "text generation"
 ---
-### Model description
 GPT-2 model from Lithuania using Wikipedia corpus dataset based on GPT-2 small model.
 This is only the first version of the model, over time model will be improved using a bigger dataset and better data preparation.
-### Training data
 This model was pre-trained with 180MB of Lithuanian Wikipedia. The texts are tokenized using a byte-level version of Byte Pair Encoding (BPE).
-### Training
 The model was trained on wiki-corpus for 40 hours using NVIDIA Tesla P100 GPU.
-### How to use
-## Load model
 ``` from transformers import AutoTokenizer, TFAutoModelWithLMHead
 import tensorflow as tf
@@ -47,11 +47,11 @@ outputs = model.generate(inputs, eos_token_id=50256, pad_token_id=50256,
 print(tokenizer.decode(outputs[0]))
 ```
-### Limitations and bias
 The training data used for this model come from Lithuanian Wikipedia. We know it contains a lot of unfiltered content from the internet, which is far from neutral. As the openAI team themselves point out in their model card:
 "Because large-scale language models like GPT-2 do not distinguish fact from fiction, we don’t support use-cases that require the generated text to be true. Additionally, language models like GPT-2 reflect the biases inherent to the systems they were trained on, so we do not recommend that they be deployed into systems that interact with humans > unless the deployers first carry out a study of biases relevant to the intended use-case. We found no statistically significant difference in gender, race, and religious bias probes between 774M and 1.5B, implying all versions of GPT-2 should be approached with similar levels of caution around use cases that are sensitive to biases around human attributes."
-### Author
 Lithuanian GPT-2 small was trained and evaluated by Deividas Mataciunas

 tags:
   - "text generation"
 ---
+## Model description
 GPT-2 model from Lithuania using Wikipedia corpus dataset based on GPT-2 small model.
 This is only the first version of the model, over time model will be improved using a bigger dataset and better data preparation.
+## Training data
 This model was pre-trained with 180MB of Lithuanian Wikipedia. The texts are tokenized using a byte-level version of Byte Pair Encoding (BPE).
+## Training
 The model was trained on wiki-corpus for 40 hours using NVIDIA Tesla P100 GPU.
+## How to use
+### Load model
 ``` from transformers import AutoTokenizer, TFAutoModelWithLMHead
 import tensorflow as tf
 print(tokenizer.decode(outputs[0]))
 ```
+## Limitations and bias
 The training data used for this model come from Lithuanian Wikipedia. We know it contains a lot of unfiltered content from the internet, which is far from neutral. As the openAI team themselves point out in their model card:
 "Because large-scale language models like GPT-2 do not distinguish fact from fiction, we don’t support use-cases that require the generated text to be true. Additionally, language models like GPT-2 reflect the biases inherent to the systems they were trained on, so we do not recommend that they be deployed into systems that interact with humans > unless the deployers first carry out a study of biases relevant to the intended use-case. We found no statistically significant difference in gender, race, and religious bias probes between 774M and 1.5B, implying all versions of GPT-2 should be approached with similar levels of caution around use cases that are sensitive to biases around human attributes."
+## Author
 Lithuanian GPT-2 small was trained and evaluated by Deividas Mataciunas