Document num_gqa parameter

It is required to be adjusted for some models, see https://github.com/jmorganca/ollama/issues/320 for more context
This commit is contained in:
Arturas Smorgun 2023-08-11 00:58:09 +01:00 committed by GitHub
parent 21e6197c0b
commit c0e7a3b90e
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23

View file

@ -104,6 +104,7 @@ PARAMETER <parameter> <parametervalue>
| mirostat_eta | Influences how quickly the algorithm responds to feedback from the generated text. A lower learning rate will result in slower adjustments, while a higher learning rate will make the algorithm more responsive. (Default: 0.1) | float | mirostat_eta 0.1 | | mirostat_eta | Influences how quickly the algorithm responds to feedback from the generated text. A lower learning rate will result in slower adjustments, while a higher learning rate will make the algorithm more responsive. (Default: 0.1) | float | mirostat_eta 0.1 |
| mirostat_tau | Controls the balance between coherence and diversity of the output. A lower value will result in more focused and coherent text. (Default: 5.0) | float | mirostat_tau 5.0 | | mirostat_tau | Controls the balance between coherence and diversity of the output. A lower value will result in more focused and coherent text. (Default: 5.0) | float | mirostat_tau 5.0 |
| num_ctx | Sets the size of the context window used to generate the next token. (Default: 2048) | int | num_ctx 4096 | | num_ctx | Sets the size of the context window used to generate the next token. (Default: 2048) | int | num_ctx 4096 |
| num_gqa | The number of GQA groups in the transformer layer. Required for some models, for example it is 8 for llama2:70b | int | num_gqa 8 |
| num_gpu | The number of GPUs to use. On macOS it defaults to 1 to enable metal support, 0 to disable. | int | num_gpu 1 | | num_gpu | The number of GPUs to use. On macOS it defaults to 1 to enable metal support, 0 to disable. | int | num_gpu 1 |
| num_thread | Sets the number of threads to use during computation. By default, Ollama will detect this for optimal performance. It is recommended to set this value to the number of physical CPU cores your system has (as opposed to the logical number of cores). | int | num_thread 8 | | num_thread | Sets the number of threads to use during computation. By default, Ollama will detect this for optimal performance. It is recommended to set this value to the number of physical CPU cores your system has (as opposed to the logical number of cores). | int | num_thread 8 |
| repeat_last_n | Sets how far back for the model to look back to prevent repetition. (Default: 64, 0 = disabled, -1 = num_ctx) | int | repeat_last_n 64 | | repeat_last_n | Sets how far back for the model to look back to prevent repetition. (Default: 64, 0 = disabled, -1 = num_ctx) | int | repeat_last_n 64 |