nicholasKluge commited on
Commit
bfb868d
1 Parent(s): 98a2046

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -4
README.md CHANGED
@@ -39,7 +39,7 @@ co2_eq_emissions:
39
  ---
40
  # Aira-2-portuguese-1B7
41
 
42
- `Aira-2` is the second version of the Aira instruction-tuned series. `Aira-2-portuguese-1B7` is an instruction-tuned model based on [BLOOM](https://huggingface.co/bigscience/bloom-1b7). The model was trained with a dataset composed of prompts and completions generated synthetically by prompting already-tuned models (ChatGPT, Llama, Open-Assistant, etc).
43
 
44
  Check our gradio-demo in [Spaces](https://huggingface.co/spaces/nicholasKluge/Aira-Demo-Portuguese).
45
 
@@ -105,9 +105,11 @@ The model will output something like:
105
 
106
  ## Limitations
107
 
108
- 🤥 Generative models can perpetuate the generation of pseudo-informative content, that is, false information that may appear truthful.
109
 
110
- 🤬 In certain types of tasks, generative models can produce harmful and discriminatory content inspired by historical stereotypes.
 
 
111
 
112
  ## Cite as 🤗
113
 
@@ -127,4 +129,4 @@ The model will output something like:
127
 
128
  ## License
129
 
130
- The `Aira-2-portuguese-1B7` is licensed under the RAIL License since it is a model derived from BLOOM. See the [LICENSE](LICENSE) file for more details.
 
39
  ---
40
  # Aira-2-portuguese-1B7
41
 
42
+ Aira-2 is the second version of the Aira instruction-tuned series. Aira-2-portuguese-1B7 is an instruction-tuned model based on [BLOOM](https://huggingface.co/bigscience/bloom-1b7). The model was trained with a dataset composed of prompts and completions generated synthetically by prompting already-tuned models (ChatGPT, Llama, Open-Assistant, etc).
43
 
44
  Check our gradio-demo in [Spaces](https://huggingface.co/spaces/nicholasKluge/Aira-Demo-Portuguese).
45
 
 
105
 
106
  ## Limitations
107
 
108
+ - **Hallucinations:** This model can produce content that can be mistaken for truth but is, in fact, misleading or entirely false, i.e., hallucination.
109
 
110
+ - **Biases and Toxicity:** This model inherits the social and historical stereotypes from the data used to train it. Given these biases, the model can produce toxic content, i.e., harmful, offensive, or detrimental to individuals, groups, or communities.
111
+
112
+ - **Repetition and Verbosity:** The model may get stuck on repetition loops (especially if the repetition penalty during generations is set to a meager value) or produce verbose responses unrelated to the prompt it was given.
113
 
114
  ## Cite as 🤗
115
 
 
129
 
130
  ## License
131
 
132
+ Aira-2-portuguese-1B7 is licensed under the RAIL License since it is a model derived from BLOOM. See the [LICENSE](LICENSE) file for more details.