vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

server: fix incorrectly reported token probabilities (#7125)

Browse source 
* server: normalize token probabilities

* fix temperature == 0.0f
This commit is contained in:
Johannes Gäßler 2024-05-07 23:07:58 +02:00 • committed by GitHub
parent b6aa670203
commit af0a5b6163
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
4 changed files with 31 additions and 11 deletions
Download patch file Download diff file Expand all files Collapse all files

2
examples/server/README.md
View file

@ -272,7 +272,7 @@ node index.js
`logit_bias`: Modify the likelihood of a token appearing in the generated text completion. For example, use `"logit_bias": [[15043,1.0]]` to increase the likelihood of the token 'Hello', or `"logit_bias": [[15043,-1.0]]` to decrease its likelihood. Setting the value to false, `"logit_bias": [[15043,false]]` ensures that the token `Hello` is never produced. The tokens can also be represented as strings, e.g. `[["Hello, World!",-0.5]]` will reduce the likelihood of all the individual tokens that represent the string `Hello, World!`, just like the `presence_penalty` does. Default: `[]`
`n_probs`: If greater than 0, the response also contains the probabilities of top N tokens for each generated token. Default: `0`
`n_probs`: If greater than 0, the response also contains the probabilities of top N tokens for each generated token given the sampling settings. Note that for temperature < 0 the tokens are sampled greedily but token probabilities are still being calculated via a simple softmax of the logits without considering any other sampler settings. Default: `0`
`min_keep`: If greater than 0, force samplers to return N possible tokens at minimum. Default: `0`