LLM: Update convert of gpenox to sync with new libgptneox.so (#8345)
This commit is contained in:
parent
50dd9dd1c5
commit
9daf543e2f
1 changed files with 6 additions and 1 deletions
|
|
@ -1276,7 +1276,12 @@ def _convert_gptneox_hf_to_ggml(model_path, outfile_dir, outtype):
|
||||||
fout.write(struct.pack("i", ftype))
|
fout.write(struct.pack("i", ftype))
|
||||||
|
|
||||||
dot_token = tokenizer.encode(".")[0]
|
dot_token = tokenizer.encode(".")[0]
|
||||||
|
vocab = tokenizer.vocab
|
||||||
|
id2token = {v: k for k, v in vocab.items()}
|
||||||
for i in range(hparams["vocab_size"]):
|
for i in range(hparams["vocab_size"]):
|
||||||
|
if i in id2token:
|
||||||
|
text = id2token[i].encode('utf-8')
|
||||||
|
else:
|
||||||
text = tokenizer.decode([i]).encode('utf-8')
|
text = tokenizer.decode([i]).encode('utf-8')
|
||||||
fout.write(struct.pack("i", len(text)))
|
fout.write(struct.pack("i", len(text)))
|
||||||
fout.write(text)
|
fout.write(text)
|
||||||
|
|
|
||||||
Loading…
Reference in a new issue