Skip to content

Commit

Permalink
llama : add gemma model (ggerganov#5631)
Browse files Browse the repository at this point in the history 
There are couple things in this architecture:

1. Shared input and output embedding parameters.
2. Key length and value length are not derived from `n_embd`.

More information about the models can be found at
https://ai.google.dev/gemma. GGUFs can be downloaded from
https://huggingface.co/google.
  • Loading branch information
@postmasters @jordankanter
postmasters authored and jordankanter committed Mar 13, 2024
1 parent 8204bf8 commit fa4b8c3
Show file tree
Hide file tree
Showing 3 changed files with 186 additions and 0 deletions.
1 change: 1 addition & 0 deletions README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -107,6 +107,7 @@ Typically finetunes of the base models below are supported as well.
- [x] [Orion 14B](https://github.com/ggerganov/llama.cpp/pull/5118)
- [x] [InternLM2](https://huggingface.co/models?search=internlm2)
- [x] [CodeShell](https://github.com/WisdomShell/codeshell)
- [x] [Gemma](https://ai.google.dev/gemma)

**Multimodal models:**

Expand Down
15 changes: 15 additions & 0 deletions gguf-py/gguf/constants.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -111,6 +111,7 @@ class MODEL_ARCH(IntEnum):
ORION = auto()
INTERNLM2 = auto()
MINICPM = auto()
GEMMA = auto()


class MODEL_TENSOR(IntEnum):
Expand Down Expand Up @@ -167,6 +168,7 @@ class MODEL_TENSOR(IntEnum):
MODEL_ARCH.ORION: "orion",
MODEL_ARCH.INTERNLM2: "internlm2",
MODEL_ARCH.MINICPM: "minicpm",
MODEL_ARCH.GEMMA: "gemma",
}

TENSOR_NAMES: dict[MODEL_TENSOR, str] = {
Expand Down Expand Up @@ -511,6 +513,19 @@ class MODEL_TENSOR(IntEnum):
MODEL_TENSOR.FFN_DOWN_EXP,
MODEL_TENSOR.FFN_UP_EXP,
],
MODEL_ARCH.GEMMA: [
MODEL_TENSOR.TOKEN_EMBD,
MODEL_TENSOR.OUTPUT_NORM,
MODEL_TENSOR.ATTN_NORM,
MODEL_TENSOR.ATTN_Q,
MODEL_TENSOR.ATTN_K,
MODEL_TENSOR.ATTN_V,
MODEL_TENSOR.ATTN_OUT,
MODEL_TENSOR.FFN_GATE,
MODEL_TENSOR.FFN_DOWN,
MODEL_TENSOR.FFN_UP,
MODEL_TENSOR.FFN_NORM,
],
# TODO
}

Expand Down
170 changes: 170 additions & 0 deletions llama.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -208,6 +208,7 @@ enum llm_arch {
LLM_ARCH_ORION,
LLM_ARCH_INTERNLM2,
LLM_ARCH_MINICPM,
LLM_ARCH_GEMMA,
LLM_ARCH_UNKNOWN,
};

Expand All @@ -234,6 +235,7 @@ static std::map<llm_arch, const char *> LLM_ARCH_NAMES = {
{ LLM_ARCH_ORION, "orion" },
{ LLM_ARCH_INTERNLM2, "internlm2" },
{ LLM_ARCH_MINICPM, "minicpm" },
{ LLM_ARCH_GEMMA, "gemma" },
};

enum llm_kv {
Expand Down Expand Up @@ -760,6 +762,22 @@ static std::map<llm_arch, std::map<llm_tensor, std::string>> LLM_TENSOR_NAMES =
{ LLM_TENSOR_FFN_UP_EXP, "blk.%d.ffn_up.%d" },
},
},
{
LLM_ARCH_GEMMA,
{
{ LLM_TENSOR_TOKEN_EMBD, "token_embd" },
{ LLM_TENSOR_OUTPUT_NORM, "output_norm" },
{ LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm" },
{ LLM_TENSOR_ATTN_Q, "blk.%d.attn_q" },
{ LLM_TENSOR_ATTN_K, "blk.%d.attn_k" },
{ LLM_TENSOR_ATTN_V, "blk.%d.attn_v" },
{ LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
{ LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
{ LLM_TENSOR_FFN_GATE, "blk.%d.ffn_gate" },
{ LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
{ LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
},
},
{
LLM_ARCH_UNKNOWN,
{
Expand Down Expand Up @@ -3243,6 +3261,16 @@ static void llm_load_hparams(
default: model.type = e_model::MODEL_UNKNOWN;
}
} break;
case LLM_ARCH_GEMMA:
{
ml.get_key(LLM_KV_ATTENTION_LAYERNORM_RMS_EPS, hparams.f_norm_rms_eps);

switch (hparams.n_layer) {
case 18: model.type = e_model::MODEL_2B; break;
case 28: model.type = e_model::MODEL_7B; break;
default: model.type = e_model::MODEL_UNKNOWN;
}
} break;
default: (void)0;
}

Expand Down Expand Up @@ -4360,6 +4388,37 @@ static bool llm_load_tensors(
layer.ffn_up = ml.create_tensor(ctx_split, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff});
}
} break;
case LLM_ARCH_GEMMA:
{
model.tok_embd = ml.create_tensor(ctx_input, tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab});

// output
model.output_norm = ml.create_tensor(ctx_output, tn(LLM_TENSOR_OUTPUT_NORM, "weight"), {n_embd});

const int64_t n_ff = hparams.n_ff;
const int64_t n_embd_head_k = hparams.n_embd_head_k;
const int64_t n_embd_k_gqa = hparams.n_embd_k_gqa();
const int64_t n_embd_v_gqa = hparams.n_embd_v_gqa();

for (uint32_t i = 0; i < n_layer; ++i) {
ggml_context * ctx_layer = ctx_for_layer(i);
ggml_context * ctx_split = ctx_for_layer_split(i);

auto & layer = model.layers[i];

layer.attn_norm = ml.create_tensor(ctx_layer, tn(LLM_TENSOR_ATTN_NORM, "weight", i), {n_embd});

layer.wq = ml.create_tensor(ctx_split, tn(LLM_TENSOR_ATTN_Q, "weight", i), {n_embd, n_embd_head_k * hparams.n_head});
layer.wk = ml.create_tensor(ctx_split, tn(LLM_TENSOR_ATTN_K, "weight", i), {n_embd, n_embd_k_gqa});
layer.wv = ml.create_tensor(ctx_split, tn(LLM_TENSOR_ATTN_V, "weight", i), {n_embd, n_embd_v_gqa});
layer.wo = ml.create_tensor(ctx_split, tn(LLM_TENSOR_ATTN_OUT, "weight", i), {n_embd_head_k * hparams.n_head, n_embd});

layer.ffn_norm = ml.create_tensor(ctx_layer, tn(LLM_TENSOR_FFN_NORM, "weight", i), {n_embd});
layer.ffn_gate = ml.create_tensor(ctx_split, tn(LLM_TENSOR_FFN_GATE, "weight", i), {n_embd, n_ff});
layer.ffn_up = ml.create_tensor(ctx_split, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff});
layer.ffn_down = ml.create_tensor(ctx_split, tn(LLM_TENSOR_FFN_DOWN, "weight", i), { n_ff, n_embd});
}
} break;
default:
throw std::runtime_error("unknown architecture");
}
Expand Down Expand Up @@ -7366,6 +7425,113 @@ struct llm_build_context {

return gf;
}

struct ggml_cgraph * build_gemma() {
struct ggml_cgraph * gf = ggml_new_graph_custom(ctx0, LLAMA_MAX_NODES, false);

const int64_t n_embd_head_k = hparams.n_embd_head_k;

struct ggml_tensor * cur;
struct ggml_tensor * inpL;

inpL = llm_build_inp_embd(ctx0, hparams, batch, model.tok_embd, lctx.inp_tokens, lctx.inp_embd, cb);
cb(inpL, "inp_embd", -1);
inpL = ggml_scale(ctx0, inpL, sqrtf(n_embd));
cb(inpL, "inp_scaled", -1);

// inp_pos - contains the positions
struct ggml_tensor * inp_pos = ggml_view_1d(ctx0, lctx.inp_pos, n_tokens, 0);
cb(inp_pos, "inp_pos", -1);

// KQ_mask (mask for 1 head, it will be broadcasted to all heads)
struct ggml_tensor * KQ_mask = ggml_view_2d(ctx0, lctx.inp_KQ_mask, n_kv, n_tokens, n_kv*ggml_type_size(lctx.inp_KQ_mask->type), 0);
cb(KQ_mask, "KQ_mask", -1);

// shift the entire K-cache if needed
if (do_rope_shift) {
llm_build_k_shift(ctx0, hparams, cparams, kv_self, gf, lctx.inp_K_shift, LLM_ROPE, n_ctx, freq_base, freq_scale, cb);
}

for (int il = 0; il < n_layer; ++il) {

// norm
cur = llm_build_norm(ctx0, inpL, hparams,
model.layers[il].attn_norm, NULL,
LLM_NORM_RMS, cb, il);
cb(cur, "attn_norm", il);

// self-attention
{
// compute Q and K and RoPE them
struct ggml_tensor * Qcur = ggml_mul_mat(ctx0, model.layers[il].wq, cur);
cb(Qcur, "Qcur", il);

struct ggml_tensor * Kcur = ggml_mul_mat(ctx0, model.layers[il].wk, cur);
cb(Kcur, "Kcur", il);

struct ggml_tensor * Vcur = ggml_mul_mat(ctx0, model.layers[il].wv, cur);
cb(Vcur, "Vcur", il);

Qcur = ggml_rope_custom(
ctx0, ggml_reshape_3d(ctx0, Qcur, n_embd_head_k, n_head, n_tokens), inp_pos,
n_embd_head_k, 2, 0, n_orig_ctx, freq_base, freq_scale,
ext_factor, attn_factor, beta_fast, beta_slow);
cb(Qcur, "Qcur", il);
Qcur = ggml_scale(ctx0, Qcur, 1.0f / sqrtf(float(n_embd_head_k)));
cb(Qcur, "Qcur_scaled", il);

Kcur = ggml_rope_custom(
ctx0, ggml_reshape_3d(ctx0, Kcur, n_embd_head_k, n_head_kv, n_tokens), inp_pos,
n_embd_head_k, 2, 0, n_orig_ctx, freq_base, freq_scale,
ext_factor, attn_factor, beta_fast, beta_slow);
cb(Kcur, "Kcur", il);

cur = llm_build_kv(ctx0, model, hparams, kv_self, gf,
model.layers[il].wo, NULL,
Kcur, Vcur, Qcur, KQ_mask, nullptr, n_ctx, n_tokens, kv_head, n_kv, 1.0f, cb, il);
cb(cur, "kqv_out", il);
}
struct ggml_tensor * sa_out = ggml_add(ctx0, cur, inpL);
cb(sa_out, "sa_out", il);

cur = llm_build_norm(ctx0, sa_out, hparams,
model.layers[il].ffn_norm, NULL,
LLM_NORM_RMS, cb, il);
cb(cur, "ffn_norm", il);

// feed-forward network
{
cur = llm_build_ffn(ctx0, cur,
model.layers[il].ffn_up, NULL,
model.layers[il].ffn_gate, NULL,
model.layers[il].ffn_down, NULL,
NULL,
LLM_FFN_GELU, LLM_FFN_PAR, cb, il);
cb(cur, "ffn_out", il);
}

cur = ggml_add(ctx0, cur, sa_out);
cb(cur, "l_out", il);

// input for next layer
inpL = cur;
}

cur = inpL;

cur = llm_build_norm(ctx0, cur, hparams,
model.output_norm, NULL,
LLM_NORM_RMS, cb, -1);
cb(cur, "result_norm", -1);

// lm_head
cur = ggml_mul_mat(ctx0, model.tok_embd, cur);
cb(cur, "result_output", -1);

ggml_build_forward_expand(gf, cur);

return gf;
}
};

static struct ggml_cgraph * llama_build_graph(
Expand Down Expand Up @@ -7474,6 +7640,10 @@ static struct ggml_cgraph * llama_build_graph(
{
result = llm.build_minicpm();
} break;
case LLM_ARCH_GEMMA:
{
result = llm.build_gemma();
} break;
default:
GGML_ASSERT(false);
}
Expand Down

0 comments on commit fa4b8c3

Please sign in to comment.