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

common.h : add hellaswag / remove perplexity-lines

Browse source
This commit is contained in:
klosax 2023-07-25 17:31:43 +02:00 • committed by GitHub
parent 07aaa0f63f
commit 522a29c426
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 4 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

6
examples/common.h
View file

@ -70,7 +70,10 @@ struct gpt_params {
std::string lora_adapter = ""; // lora adapter path std::string lora_adapter = ""; // lora adapter path
std::string lora_base = ""; // base model path for the lora adapter std::string lora_base = ""; // base model path for the lora adapter
bool low_vram = false; // if true, reduce VRAM usage at the cost of performance bool hellaswag = false; // compute HellaSwag score over random examples from datafile supplied in prompt
size_t hellaswag_tasks = 200; // number of examples to use when computing the HellaSwag score
bool low_vram = false; // if true, reduce VRAM usage at the cost of performance
bool memory_f16 = true; // use f16 instead of f32 for memory kv bool memory_f16 = true; // use f16 instead of f32 for memory kv
bool random_prompt = false; // do not randomize prompt if none provided bool random_prompt = false; // do not randomize prompt if none provided
bool use_color = false; // use color to distinguish generations and inputs bool use_color = false; // use color to distinguish generations and inputs
@ -86,7 +89,6 @@ struct gpt_params {
bool instruct = false; // instruction mode (used for Alpaca models) bool instruct = false; // instruction mode (used for Alpaca models)
bool penalize_nl = true; // consider newlines as a repeatable token bool penalize_nl = true; // consider newlines as a repeatable token
bool perplexity = false; // compute perplexity over the prompt bool perplexity = false; // compute perplexity over the prompt
bool perplexity_lines = false; // compute perplexity over each line of the prompt
bool use_mmap = true; // use mmap for faster loads bool use_mmap = true; // use mmap for faster loads
bool use_mlock = false; // use mlock to keep model in memory bool use_mlock = false; // use mlock to keep model in memory
bool mem_test = false; // compute maximum memory usage bool mem_test = false; // compute maximum memory usage