Skip to content

Commit

Permalink
ggml : fix q2_k bpw in comments (ggml/680)
Browse files Browse the repository at this point in the history
  • Loading branch information
ggerganov committed Jan 5, 2024
1 parent 1bf681f commit d061bf9
Showing 1 changed file with 1 addition and 1 deletion.
2 changes: 1 addition & 1 deletion ggml-quants.h
Original file line number Diff line number Diff line change
Expand Up @@ -70,7 +70,7 @@ static_assert(sizeof(block_q8_1) == 2*sizeof(float) + QK8_1, "wrong q8_1 block s
// 2-bit quantization
// weight is represented as x = a * q + b
// 16 blocks of 16 elements each
// Effectively 2.5625 bits per weight
// Effectively 2.625 bits per weight
typedef struct {
uint8_t scales[QK_K/16]; // scales and mins, quantized with 4 bits
uint8_t qs[QK_K/4]; // quants
Expand Down

0 comments on commit d061bf9

Please sign in to comment.