为了安全，强烈建议开启2FA双因子认证：User Settings -> Account -> Enable two-factor authentication！！！

Tags

Tags give the ability to mark specific points in history as being important

b1539

e9370664 · gguf-py : export chat templates (#4125) · Nov 19, 2023
b1538

28a2e6e7 · tokenize example: Respect normal add BOS token behavior (#4126) · Nov 18, 2023
b1536

2923f17f · Clean up ggml-cuda.cu warnings when compiling with clang (for ROCM) (#4124) · Nov 18, 2023
b1535

bbecf3f4 · llama : increase max nodes (#4115) · Nov 17, 2023
b1534

8e936108 · build : support ppc64le build for make and CMake (#3963) · Nov 17, 2023
b1533

5ad387e9 · tokenize : fix trailing whitespace · Nov 17, 2023
b1532

2fa02b4b · examples : add tokenize (#4039) · Nov 17, 2023
b1529

9e87ef60 · common : improve yaml log escaping (#4080) · Nov 17, 2023
b1528

c7cce124 · llava : fix compilation warning that fread return value is not used (#4069) · Nov 17, 2023
b1526

ba4cf5c0 · train : move number of gpu layers argument parsing to common/train.cpp (#4074) · Nov 17, 2023
b1525

e85bb1a8 · llama : add functions to get the model's metadata (#4013) · Nov 17, 2023
b1524

3e916a07 · finetune : speed-up ggml_compute_forward_out_prod_f32 via BLAS (#4079) · Nov 17, 2023
b1523

947f64f1 · finetune : zero the loraB initial vectors (#4082) · Nov 17, 2023
b1522

b83e149e · cuda : get_row_rounding F32 (#4095) · Nov 17, 2023
b1521

4f447a48 · llama : fix data units (#4101) · Nov 17, 2023
b1520

91f64993 · Respect tokenizer.ggml.add_bos_token value when tokenizing (#4040) · Nov 16, 2023
b1519

8da46278 · gguf : fix potential infinite loops while parsing (#4100) · Nov 16, 2023
b1518

a6fc554e · llama : restore prefix space in llama tokenizer (#4081) · Nov 15, 2023
b1517

1cf2850d · ggml-cuda : increase max graph size (#4084) · Nov 15, 2023
b1516

6bb4908a · Fix MacOS Sonoma model quantization (#4052) · Nov 14, 2023