Files changed (1) hide show
  1. README.md +4 -4
README.md CHANGED
@@ -18,7 +18,7 @@ Our model hasn't been fine-tuned through reinforcement learning from human feedb
18
 
19
  ## Intended Uses
20
 
21
- Phi-2 is intended for research purposes only. Given the nature of the training data, the Phi-2 model is best suited for prompts using the QA format, the chat format, and the code format.
22
 
23
  ### QA Format:
24
 
@@ -68,8 +68,8 @@ def print_prime(n):
68
  where the model generates the text after the comments.
69
 
70
  **Notes:**
71
- * Phi-2 is intended for research purposes. The model-generated text/code should be treated as a starting point rather than a definitive solution for potential use cases. Users should be cautious when employing these models in their applications.
72
- * Direct adoption for production tasks is out of the scope of this research project. As a result, the Phi-2 model has not been tested to ensure that it performs adequately for any production-level application. Please refer to the limitation sections of this document for more details.
73
  * If you are using `transformers>=4.36.0`, always load the model with `trust_remote_code=True` to prevent side-effects.
74
 
75
  ## Sample Code
@@ -129,7 +129,7 @@ Furthermore, in the forward pass of the model, we currently do not support outpu
129
 
130
  * Potential Societal Biases: Phi-2 is not entirely free from societal biases despite efforts in assuring trainig data safety. There's a possibility it may generate content that mirrors these societal biases, particularly if prompted or instructed to do so. We urge users to be aware of this and to exercise caution and critical thinking when interpreting model outputs.
131
 
132
- * Toxicity: Despite being trained with carefully selected data, the model can still produce harmful content if explicitly prompted or instructed to do so. We chose to release the model for research purposes only -- We hope to help the open-source community develop the most effective ways to reduce the toxicity of a model directly after pretraining.
133
 
134
  * Verbosity: Phi-2 being a base model often produces irrelevant or extra text and responses following its first answer to user prompts within a single turn. This is due to its training dataset being primarily textbooks, which results in textbook-like responses.
135
 
 
18
 
19
  ## Intended Uses
20
 
21
+ Given the nature of the training data, the Phi-2 model is best suited for prompts using the QA format, the chat format, and the code format.
22
 
23
  ### QA Format:
24
 
 
68
  where the model generates the text after the comments.
69
 
70
  **Notes:**
71
+ * Phi-2 is intended for QA, chat, and code purposes. The model-generated text/code should be treated as a starting point rather than a definitive solution for potential use cases. Users should be cautious when employing these models in their applications.
72
+ * Direct adoption for production tasks without evaluation is out of scope of this project. As a result, the Phi-2 model has not been tested to ensure that it performs adequately for any production-level application. Please refer to the limitation sections of this document for more details.
73
  * If you are using `transformers>=4.36.0`, always load the model with `trust_remote_code=True` to prevent side-effects.
74
 
75
  ## Sample Code
 
129
 
130
  * Potential Societal Biases: Phi-2 is not entirely free from societal biases despite efforts in assuring trainig data safety. There's a possibility it may generate content that mirrors these societal biases, particularly if prompted or instructed to do so. We urge users to be aware of this and to exercise caution and critical thinking when interpreting model outputs.
131
 
132
+ * Toxicity: Despite being trained with carefully selected data, the model can still produce harmful content if explicitly prompted or instructed to do so. We chose to release the model to help the open-source community develop the most effective ways to reduce the toxicity of a model directly after pretraining.
133
 
134
  * Verbosity: Phi-2 being a base model often produces irrelevant or extra text and responses following its first answer to user prompts within a single turn. This is due to its training dataset being primarily textbooks, which results in textbook-like responses.
135