model : Granite MoE shared (#13269)

* feat: Add GGUF conversion for granitemoeshared

Branch: GraniteMoEShared

Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>

* feat: hparam and arch plumbing for granitemoeshared

Branch: GraniteMoEShared

Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>

* fix: Split MoE fused tensors for shared experts in conversion

Branch: GraniteMoEShared

Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>

* feat: First WIP cut at model arch in cpp

The hparam and architecture plumbing should be correct, but the
implementation of the shared experts seems to still be broken.

Branch: GraniteMoEShared

Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>

* fix: Cleaner (maybe more correct?) splitting for gate/up

Branch: GraniteMoEShared

Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>

* fix: Fix the input to the shared experts

I had misread that the shared experts take the inputs _before_ the standard
MoE layer and was feeding the output of the MoE to the shared experts.

Branch: GraniteMoEShared

Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>

* fix: Avoid architecture-specific checks for Granite MoE Shared

This is a cleaner way that will allow more flexibility in architecture
strings going forward.

Branch: GraniteMoEShared

Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>

* refactor: Split granite architectures out of llm_build_llama

This helps de-clutter the llama-family graph construction and allows
granite to diverge further (in preparation for Granite 4).

NOTE: I removed the granite scale factors from llm_build_deci because they
appear to only be there as copy-paste from llm_build_llama. The HF config
does not seem to set those values:
https://huggingface.co/Deci/DeciLM-7B/blob/main/config.json

Branch: GraniteMoEShared

Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>

* fix: Fix compiler warning about uninitialized inp_pos

This should not have been reachable, but it warns on some compliers

Branch: GraniteMoEShared

Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>

* fix: Consoladate GraniteMoEShared into GraniteMoE for conversion

Branch: GraniteMoEShared

Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>

* fix: Consolidate GraniteMoEShared into GraniteMoE on the c++ side

Branch: GraniteMoEShared

Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>

---------

Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>
This commit is contained in:
Gabe Goodhart
2025-05-13 07:12:01 -06:00
committed by GitHub
parent 1e2809bc4b
commit d590cd4c24
5 changed files with 235 additions and 35 deletions

View File

@ -5746,11 +5746,20 @@ class GraniteModel(LlamaModel):
logger.info("gguf: (granite) logits_scale = %s", logits_scale) logger.info("gguf: (granite) logits_scale = %s", logits_scale)
@ModelBase.register("GraniteMoeForCausalLM") @ModelBase.register("GraniteMoeForCausalLM", "GraniteMoeSharedForCausalLM")
class GraniteMoeModel(GraniteModel): class GraniteMoeModel(GraniteModel):
"""Conversion for IBM's GraniteMoeForCausalLM""" """Conversion for IBM's GraniteMoeForCausalLM"""
model_arch = gguf.MODEL_ARCH.GRANITE_MOE model_arch = gguf.MODEL_ARCH.GRANITE_MOE
def set_gguf_parameters(self):
"""GraniteMoeShared uses GraniteMoe parameters plus the following:
- shared_intermediate_size
"""
super().set_gguf_parameters()
if shared_feed_forward_length := self.hparams.get("shared_intermediate_size"):
self.gguf_writer.add_expert_shared_feed_forward_length(shared_feed_forward_length)
logger.info("gguf: (granitemoeshared) shared_feed_forward_length = %s", shared_feed_forward_length)
def modify_tensors(self, data_torch: Tensor, name: str, bid: int | None) -> Iterable[tuple[str, Tensor]]: def modify_tensors(self, data_torch: Tensor, name: str, bid: int | None) -> Iterable[tuple[str, Tensor]]:
"""In modeling_granitemoe, the JetMoe implementation of parallel experts """In modeling_granitemoe, the JetMoe implementation of parallel experts
is used. This essentially merges w1 and w3 into a single tensor with 2x is used. This essentially merges w1 and w3 into a single tensor with 2x
@ -5761,12 +5770,21 @@ class GraniteMoeModel(GraniteModel):
if name.endswith("block_sparse_moe.input_linear.weight"): if name.endswith("block_sparse_moe.input_linear.weight"):
ffn_dim = self.hparams["intermediate_size"] ffn_dim = self.hparams["intermediate_size"]
assert data_torch.shape[-2] == 2 * ffn_dim, "Merged FFN tensor size must be 2 * intermediate_size" assert data_torch.shape[-2] == 2 * ffn_dim, "Merged FFN tensor size must be 2 * intermediate_size"
gate, up = data_torch[..., :ffn_dim, :], data_torch[..., ffn_dim:, :] gate, up = data_torch.split(ffn_dim, dim=-2)
return [ return [
(self.format_tensor_name(gguf.MODEL_TENSOR.FFN_GATE_EXP, bid), gate), (self.format_tensor_name(gguf.MODEL_TENSOR.FFN_GATE_EXP, bid), gate),
(self.format_tensor_name(gguf.MODEL_TENSOR.FFN_UP_EXP, bid), up), (self.format_tensor_name(gguf.MODEL_TENSOR.FFN_UP_EXP, bid), up),
] ]
if name.endswith("shared_mlp.input_linear.weight"):
ffn_dim = self.hparams["shared_intermediate_size"]
assert data_torch.shape[-2] == 2 * ffn_dim, "Merged FFN tensor size must be 2 * shared_intermediate_size"
gate, up = data_torch.split(ffn_dim, dim=-2)
return [
(self.format_tensor_name(gguf.MODEL_TENSOR.FFN_GATE_SHEXP, bid), gate),
(self.format_tensor_name(gguf.MODEL_TENSOR.FFN_UP_SHEXP, bid), up),
]
return super().modify_tensors(data_torch, name, bid) return super().modify_tensors(data_torch, name, bid)

View File

@ -1905,6 +1905,9 @@ MODEL_TENSORS: dict[MODEL_ARCH, list[MODEL_TENSOR]] = {
MODEL_TENSOR.FFN_GATE_EXP, MODEL_TENSOR.FFN_GATE_EXP,
MODEL_TENSOR.FFN_DOWN_EXP, MODEL_TENSOR.FFN_DOWN_EXP,
MODEL_TENSOR.FFN_UP_EXP, MODEL_TENSOR.FFN_UP_EXP,
MODEL_TENSOR.FFN_GATE_SHEXP,
MODEL_TENSOR.FFN_UP_SHEXP,
MODEL_TENSOR.FFN_DOWN_SHEXP,
], ],
MODEL_ARCH.CHAMELEON: [ MODEL_ARCH.CHAMELEON: [
MODEL_TENSOR.TOKEN_EMBD, MODEL_TENSOR.TOKEN_EMBD,

View File

@ -428,6 +428,7 @@ class TensorNameMap:
"model.layers.{bid}.mlp.shared_expert.down_proj", # qwen2moe "model.layers.{bid}.mlp.shared_expert.down_proj", # qwen2moe
"model.layers.{bid}.mlp.shared_experts.down_proj", # deepseek deepseek2 "model.layers.{bid}.mlp.shared_experts.down_proj", # deepseek deepseek2
"language_model.model.layers.{bid}.feed_forward.shared_expert.down_proj", # llama4 "language_model.model.layers.{bid}.feed_forward.shared_expert.down_proj", # llama4
"model.layers.{bid}.shared_mlp.output_linear", # granitemoe
), ),
MODEL_TENSOR.ATTN_Q_NORM: ( MODEL_TENSOR.ATTN_Q_NORM: (

View File

@ -1481,6 +1481,9 @@ static const std::map<llm_arch, std::map<llm_tensor, const char *>> LLM_TENSOR_N
{ LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" }, { LLM_TENSOR_FFN_GATE_EXPS, "blk.%d.ffn_gate_exps" },
{ LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" }, { LLM_TENSOR_FFN_DOWN_EXPS, "blk.%d.ffn_down_exps" },
{ LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" }, { LLM_TENSOR_FFN_UP_EXPS, "blk.%d.ffn_up_exps" },
{ LLM_TENSOR_FFN_GATE_SHEXP, "blk.%d.ffn_gate_shexp" },
{ LLM_TENSOR_FFN_DOWN_SHEXP, "blk.%d.ffn_down_shexp" },
{ LLM_TENSOR_FFN_UP_SHEXP, "blk.%d.ffn_up_shexp" },
}, },
}, },
{ {

View File

@ -1389,6 +1389,9 @@ void llama_model::load_hparams(llama_model_loader & ml) {
// Add additional layer/vocab/etc checks here for other model sizes // Add additional layer/vocab/etc checks here for other model sizes
default: type = LLM_TYPE_UNKNOWN; default: type = LLM_TYPE_UNKNOWN;
} }
// For Granite MoE Shared
ml.get_key(LLM_KV_EXPERT_SHARED_FEED_FORWARD_LENGTH, hparams.n_ff_shexp, /* required */ false);
} break; } break;
case LLM_ARCH_CHAMELEON: case LLM_ARCH_CHAMELEON:
{ {
@ -1772,6 +1775,13 @@ bool llama_model::load_tensors(llama_model_loader & ml) {
layer.ffn_gate_exps = create_tensor(tn(LLM_TENSOR_FFN_GATE_EXPS, "weight", i), {n_embd, n_ff, n_expert}, TENSOR_NOT_REQUIRED); layer.ffn_gate_exps = create_tensor(tn(LLM_TENSOR_FFN_GATE_EXPS, "weight", i), {n_embd, n_ff, n_expert}, TENSOR_NOT_REQUIRED);
layer.ffn_down_exps = create_tensor(tn(LLM_TENSOR_FFN_DOWN_EXPS, "weight", i), { n_ff, n_embd, n_expert}, 0); layer.ffn_down_exps = create_tensor(tn(LLM_TENSOR_FFN_DOWN_EXPS, "weight", i), { n_ff, n_embd, n_expert}, 0);
layer.ffn_up_exps = create_tensor(tn(LLM_TENSOR_FFN_UP_EXPS, "weight", i), {n_embd, n_ff, n_expert}, 0); layer.ffn_up_exps = create_tensor(tn(LLM_TENSOR_FFN_UP_EXPS, "weight", i), {n_embd, n_ff, n_expert}, 0);
// For Granite MoE Shared
if (hparams.n_ff_shexp > 0) {
layer.ffn_gate_shexp = create_tensor(tn(LLM_TENSOR_FFN_GATE_SHEXP, "weight", i), {n_embd, hparams.n_ff_shexp}, 0);
layer.ffn_up_shexp = create_tensor(tn(LLM_TENSOR_FFN_UP_SHEXP, "weight", i), {n_embd, hparams.n_ff_shexp}, 0);
layer.ffn_down_shexp = create_tensor(tn(LLM_TENSOR_FFN_DOWN_SHEXP, "weight", i), {hparams.n_ff_shexp, n_embd}, 0);
}
} }
} }
} break; } break;
@ -4385,10 +4395,13 @@ void llama_model::print_info() const {
LLAMA_LOG_INFO("%s: n_ff_exp = %d\n", __func__, hparams.n_ff_exp); LLAMA_LOG_INFO("%s: n_ff_exp = %d\n", __func__, hparams.n_ff_exp);
} }
if (arch == LLM_ARCH_MINICPM || arch == LLM_ARCH_GRANITE || arch == LLM_ARCH_GRANITE_MOE) { if (arch == LLM_ARCH_MINICPM ||
arch == LLM_ARCH_GRANITE ||
arch == LLM_ARCH_GRANITE_MOE) {
LLAMA_LOG_INFO("%s: f_embedding_scale = %f\n", __func__, hparams.f_embedding_scale); LLAMA_LOG_INFO("%s: f_embedding_scale = %f\n", __func__, hparams.f_embedding_scale);
LLAMA_LOG_INFO("%s: f_residual_scale = %f\n", __func__, hparams.f_residual_scale); LLAMA_LOG_INFO("%s: f_residual_scale = %f\n", __func__, hparams.f_residual_scale);
LLAMA_LOG_INFO("%s: f_attention_scale = %f\n", __func__, hparams.f_attention_scale); LLAMA_LOG_INFO("%s: f_attention_scale = %f\n", __func__, hparams.f_attention_scale);
LLAMA_LOG_INFO("%s: n_ff_shexp = %d\n", __func__, hparams.n_ff_shexp);
} }
if (arch == LLM_ARCH_BAILINGMOE) { if (arch == LLM_ARCH_BAILINGMOE) {
@ -4598,11 +4611,6 @@ struct llm_build_llama : public llm_graph_context {
inpSA = ggml_get_rows(ctx0, inpSA, inp_out_ids); inpSA = ggml_get_rows(ctx0, inpSA, inp_out_ids);
} }
// For Granite architecture
if (hparams.f_residual_scale) {
cur = ggml_scale(ctx0, cur, hparams.f_residual_scale);
}
ggml_tensor * ffn_inp = ggml_add(ctx0, cur, inpSA); ggml_tensor * ffn_inp = ggml_add(ctx0, cur, inpSA);
cb(ffn_inp, "ffn_inp", il); cb(ffn_inp, "ffn_inp", il);
@ -4674,11 +4682,6 @@ struct llm_build_llama : public llm_graph_context {
cb(cur, "ffn_moe_out", il); cb(cur, "ffn_moe_out", il);
} }
// For Granite architecture
if (hparams.f_residual_scale) {
cur = ggml_scale(ctx0, cur, hparams.f_residual_scale);
}
cur = ggml_add(ctx0, cur, ffn_inp); cur = ggml_add(ctx0, cur, ffn_inp);
cb(cur, "ffn_out", il); cb(cur, "ffn_out", il);
@ -4701,11 +4704,6 @@ struct llm_build_llama : public llm_graph_context {
// lm_head // lm_head
cur = build_lora_mm(model.output, cur); cur = build_lora_mm(model.output, cur);
// For Granite architecture
if (hparams.f_logit_scale) {
cur = ggml_scale(ctx0, cur, 1.0f / hparams.f_logit_scale);
}
cb(cur, "result_output", -1); cb(cur, "result_output", -1);
res->t_logits = cur; res->t_logits = cur;
@ -4816,11 +4814,6 @@ struct llm_build_deci : public llm_graph_context {
continue; continue;
} }
// For Granite architecture
if (hparams.f_residual_scale) {
cur = ggml_scale(ctx0, cur, hparams.f_residual_scale);
}
// modified to support attention-free layer of Llama-3_1-Nemotron-51B // modified to support attention-free layer of Llama-3_1-Nemotron-51B
ggml_tensor * ffn_inp = cur; ggml_tensor * ffn_inp = cur;
if (n_head > 0) { if (n_head > 0) {
@ -4844,11 +4837,6 @@ struct llm_build_deci : public llm_graph_context {
cb(cur, "ffn_out", il); cb(cur, "ffn_out", il);
} }
// For Granite architecture
if (hparams.f_residual_scale) {
cur = ggml_scale(ctx0, cur, hparams.f_residual_scale);
}
cur = ggml_add(ctx0, cur, ffn_inp); cur = ggml_add(ctx0, cur, ffn_inp);
cb(cur, "ffn_out", il); cb(cur, "ffn_out", il);
@ -4871,11 +4859,6 @@ struct llm_build_deci : public llm_graph_context {
// lm_head // lm_head
cur = build_lora_mm(model.output, cur); cur = build_lora_mm(model.output, cur);
// For Granite architecture
if (hparams.f_logit_scale) {
cur = ggml_scale(ctx0, cur, 1.0f / hparams.f_logit_scale);
}
cb(cur, "result_output", -1); cb(cur, "result_output", -1);
res->t_logits = cur; res->t_logits = cur;
@ -12214,6 +12197,195 @@ struct llm_build_arwkv7 : public llm_build_rwkv7_base {
} }
}; };
struct llm_build_granite : public llm_graph_context {
llm_build_granite(
const llama_model & model,
const llm_graph_params & params,
ggml_cgraph * gf,
const bool use_rope = true)
: llm_graph_context(params) {
const int64_t n_embd_head = hparams.n_embd_head_v;
GGML_ASSERT(n_embd_head == hparams.n_embd_head_k);
GGML_ASSERT(n_embd_head == hparams.n_rot);
ggml_tensor * cur;
ggml_tensor * inpL;
inpL = build_inp_embd(model.tok_embd);
// inp_pos - built only if rope enabled
ggml_tensor * inp_pos = nullptr;
auto * inp_attn = build_attn_inp_kv_unified();
const float kq_scale = hparams.f_attention_scale == 0.0f ? 1.0f/sqrtf(float(n_embd_head)) : hparams.f_attention_scale;
for (int il = 0; il < n_layer; ++il) {
ggml_tensor * inpSA = inpL;
// norm
cur = build_norm(inpL,
model.layers[il].attn_norm, NULL,
LLM_NORM_RMS, il);
cb(cur, "attn_norm", il);
// self-attention
{
// compute Q and K and (optionally) RoPE them
ggml_tensor * Qcur = build_lora_mm(model.layers[il].wq, cur);
cb(Qcur, "Qcur", il);
if (model.layers[il].bq) {
Qcur = ggml_add(ctx0, Qcur, model.layers[il].bq);
cb(Qcur, "Qcur", il);
}
ggml_tensor * Kcur = build_lora_mm(model.layers[il].wk, cur);
cb(Kcur, "Kcur", il);
if (model.layers[il].bk) {
Kcur = ggml_add(ctx0, Kcur, model.layers[il].bk);
cb(Kcur, "Kcur", il);
}
ggml_tensor * Vcur = build_lora_mm(model.layers[il].wv, cur);
cb(Vcur, "Vcur", il);
if (model.layers[il].bv) {
Vcur = ggml_add(ctx0, Vcur, model.layers[il].bv);
cb(Vcur, "Vcur", il);
}
Qcur = ggml_reshape_3d(ctx0, Qcur, n_embd_head, n_head, n_tokens);
Kcur = ggml_reshape_3d(ctx0, Kcur, n_embd_head, n_head_kv, n_tokens);
Vcur = ggml_reshape_3d(ctx0, Vcur, n_embd_head, n_head_kv, n_tokens);
if (use_rope) {
if (!inp_pos) {
inp_pos = build_inp_pos();
}
ggml_tensor * rope_factors = model.get_rope_factors(n_ctx_per_seq, il);
Qcur = ggml_rope_ext(
ctx0, Qcur, inp_pos, rope_factors,
n_rot, rope_type, n_ctx_orig, freq_base, freq_scale,
ext_factor, attn_factor, beta_fast, beta_slow
);
Kcur = ggml_rope_ext(
ctx0, Kcur, inp_pos, rope_factors,
n_rot, rope_type, n_ctx_orig, freq_base, freq_scale,
ext_factor, attn_factor, beta_fast, beta_slow
);
}
cb(Qcur, "Qcur", il);
cb(Kcur, "Kcur", il);
cb(Vcur, "Vcur", il);
cur = build_attn(inp_attn, gf,
model.layers[il].wo, model.layers[il].bo,
Qcur, Kcur, Vcur, nullptr, nullptr, kq_scale, il);
cb(cur, "attn_out", il);
}
if (il == n_layer - 1) {
// skip computing output for unused tokens
ggml_tensor * inp_out_ids = build_inp_out_ids();
cur = ggml_get_rows(ctx0, cur, inp_out_ids);
inpSA = ggml_get_rows(ctx0, inpSA, inp_out_ids);
}
// For Granite architectures - scale residual
cur = ggml_scale(ctx0, cur, hparams.f_residual_scale);
ggml_tensor * ffn_inp = ggml_add(ctx0, cur, inpSA);
cb(ffn_inp, "ffn_inp", il);
// feed-forward network (non-MoE)
if (model.layers[il].ffn_gate_inp == nullptr) {
cur = build_norm(ffn_inp,
model.layers[il].ffn_norm, NULL,
LLM_NORM_RMS, il);
cb(cur, "ffn_norm", il);
cur = build_ffn(cur,
model.layers[il].ffn_up, model.layers[il].ffn_up_b, NULL,
model.layers[il].ffn_gate, model.layers[il].ffn_gate_b, NULL,
model.layers[il].ffn_down, model.layers[il].ffn_down_b, NULL,
NULL,
LLM_FFN_SILU, LLM_FFN_PAR, il);
cb(cur, "ffn_out", il);
} else {
// MoE branch
cur = build_norm(ffn_inp,
model.layers[il].ffn_norm, NULL,
LLM_NORM_RMS, il);
cb(cur, "ffn_norm", il);
ggml_tensor * moe_out = build_moe_ffn(cur,
model.layers[il].ffn_gate_inp,
model.layers[il].ffn_up_exps,
model.layers[il].ffn_gate_exps,
model.layers[il].ffn_down_exps,
nullptr,
n_expert, n_expert_used,
LLM_FFN_SILU, true,
false, 0.0,
LLAMA_EXPERT_GATING_FUNC_TYPE_SOFTMAX,
il);
cb(moe_out, "ffn_moe_out", il);
// For Granite MoE Shared
if (hparams.n_ff_shexp > 0) {
ggml_tensor * ffn_shexp = build_ffn(cur,
model.layers[il].ffn_up_shexp, NULL, NULL,
model.layers[il].ffn_gate_shexp, NULL, NULL,
model.layers[il].ffn_down_shexp, NULL, NULL,
NULL,
LLM_FFN_SILU, LLM_FFN_PAR, il);
cb(ffn_shexp, "ffn_shexp", il);
cur = ggml_add(ctx0, moe_out, ffn_shexp);
cb(cur, "ffn_out", il);
} else {
cur = moe_out;
}
}
// For Granite architectures - scale residual
cur = ggml_scale(ctx0, cur, hparams.f_residual_scale);
cur = ggml_add(ctx0, cur, ffn_inp);
cb(cur, "ffn_out", il);
cur = build_cvec(cur, il);
cb(cur, "l_out", il);
// input for next layer
inpL = cur;
}
cur = inpL;
cur = build_norm(cur,
model.output_norm, NULL,
LLM_NORM_RMS, -1);
cb(cur, "result_norm", -1);
res->t_embd = cur;
// lm_head
cur = build_lora_mm(model.output, cur);
// For Granite architectures - scale logits
cur = ggml_scale(ctx0, cur, 1.0f / hparams.f_logit_scale);
cb(cur, "result_output", -1);
res->t_logits = cur;
ggml_build_forward_expand(gf, cur);
}
};
// ref: https://github.com/facebookresearch/chameleon // ref: https://github.com/facebookresearch/chameleon
// based on the original build_llama() function, changes: // based on the original build_llama() function, changes:
// * qk-norm // * qk-norm
@ -12921,8 +13093,6 @@ llm_graph_result_ptr llama_model::build_graph(
case LLM_ARCH_LLAMA: case LLM_ARCH_LLAMA:
case LLM_ARCH_LLAMA4: case LLM_ARCH_LLAMA4:
case LLM_ARCH_MINICPM: case LLM_ARCH_MINICPM:
case LLM_ARCH_GRANITE:
case LLM_ARCH_GRANITE_MOE:
{ {
llm = std::make_unique<llm_build_llama>(*this, params, gf); llm = std::make_unique<llm_build_llama>(*this, params, gf);
} break; } break;
@ -13153,6 +13323,11 @@ llm_graph_result_ptr llama_model::build_graph(
{ {
llm = std::make_unique<llm_build_arwkv7>(*this, params, gf); llm = std::make_unique<llm_build_arwkv7>(*this, params, gf);
} break; } break;
case LLM_ARCH_GRANITE:
case LLM_ARCH_GRANITE_MOE:
{
llm = std::make_unique<llm_build_granite>(*this, params, gf);
} break;
case LLM_ARCH_CHAMELEON: case LLM_ARCH_CHAMELEON:
{ {
llm = std::make_unique<llm_build_chameleon>(*this, params, gf); llm = std::make_unique<llm_build_chameleon>(*this, params, gf);