mirror of
https://github.com/ggerganov/llama.cpp.git
synced 2024-11-25 09:29:23 +01:00
Updated Tensor Encoding Schemes (markdown)
parent
3672234405
commit
2289ec20c9
@ -56,3 +56,43 @@ This is not definitive, but is helpful when reading sourcecode or console output
|
||||
| IQ4_XS | GGML_FTYPE_MOSTLY_IQ4_XS | GGML_TYPE_IQ4_XS | 4.25 | i-quantization | Superblocks has 8 blocks ( 32 weights per block) | w = func(superblock_scale, importance_matrix) | [llama.cpp PR: IQ4_XS: a 4.25 bpw quantization #5747](https://github.com/ggerganov/llama.cpp/pull/5747) |
|
||||
|
||||
* All superblocks have fp16 scaling factor and contains up to 256 weights. Number of weights in a block must be divisible by 256. (To be confirmed)
|
||||
|
||||
## Where to find the structure of these tensors in the code?
|
||||
|
||||
You would find it all usually in `ggml-common.h` where it typically be of this form
|
||||
|
||||
### Blocks
|
||||
|
||||
```c
|
||||
#define QK4_0 32
|
||||
typedef struct {
|
||||
ggml_half d; // delta
|
||||
uint8_t qs[QK4_0 / 2]; // nibbles / quants
|
||||
} block_q4_0;
|
||||
static_assert(sizeof(block_q4_0) == sizeof(ggml_half) + QK4_0 / 2, "wrong q4_0 block size/padding");
|
||||
```
|
||||
|
||||
### Superblocks
|
||||
|
||||
```c
|
||||
//
|
||||
// Super-block quantization structures
|
||||
//
|
||||
|
||||
// 2-bit quantization
|
||||
// weight is represented as x = a * q + b
|
||||
// 16 blocks of 16 elements each
|
||||
// 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
|
||||
union {
|
||||
struct {
|
||||
ggml_half d; // super-block scale for quantized scales
|
||||
ggml_half dmin; // super-block scale for quantized mins
|
||||
} GGML_COMMON_AGGR;
|
||||
ggml_half2 dm;
|
||||
};
|
||||
} block_q2_K;
|
||||
static_assert(sizeof(block_q2_K) == 2*sizeof(ggml_half) + QK_K/16 + QK_K/4, "wrong q2_K block size/padding");
|
||||
```
|
Loading…
Reference in New Issue
Block a user