|
pulsatrix 1.1.0
|
A generated continuation. More...
#include <generation.hpp>
Public Attributes | |
| std::vector< int64_t > | tokens |
| The prompt followed by the new tokens. | |
| std::vector< int64_t > | new_tokens |
| Only the new tokens, EOS included if generated. | |
| std::vector< float > | log_probs |
| Each new token's log-probability under the model itself (log-softmax of the raw logits, before any processing): how likely the model found it, whatever the decoding settings were. | |
| FinishReason | finish_reason = FinishReason::Length |
A generated continuation.
| FinishReason pulsatrix::GenerationResult::finish_reason = FinishReason::Length |
| std::vector<float> pulsatrix::GenerationResult::log_probs |
Each new token's log-probability under the model itself (log-softmax of the raw logits, before any processing): how likely the model found it, whatever the decoding settings were.
| std::vector<int64_t> pulsatrix::GenerationResult::new_tokens |
Only the new tokens, EOS included if generated.
| std::vector<int64_t> pulsatrix::GenerationResult::tokens |
The prompt followed by the new tokens.