You signed in with another tab or window.Reload to refresh your session.You signed out in another tab or window.Reload to refresh your session.You switched accounts on another tab or window.Reload to refresh your session.Dismiss alert
// What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic
53
-
Temperaturefloat32`json:"temperature,omitempty"`
53
+
Temperature*float32`json:"temperature,omitempty"`
54
54
55
55
// An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.
56
-
TopPfloat32`json:"top_p,omitempty"`
56
+
TopP*float32`json:"top_p,omitempty"`
57
57
58
58
// Number of responses to generate
59
-
Nint`json:"n,omitempty"`
59
+
N*int`json:"n,omitempty"`
60
60
61
61
// Whether or not to stream responses back as they are generated
62
62
Streambool`json:"stream,omitempty"`
@@ -68,10 +68,10 @@ type ChatCompletionRequest struct {
68
68
MaxTokensint`json:"max_tokens,omitempty"`
69
69
70
70
// (-2, 2) Penalize tokens that haven't appeared yet in the history.