models : Attach Granite NVFP4 weight scales (#31449)

* models : Attach Granite NVFP4 weight scales

* feat: Pass nvfp4 scales through build_ffn for all granite architectures

Branch: granite-nvfp4-moe-scales
AI-usage: full (Claude + Sonnet 5.5)
Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>

---------

Co-authored-by: Gabe Goodhart <ghart@us.ibm.com>
This commit is contained in:
jjperry69andGabe Goodhart authored and GitHub committed 2026-11-09 13:04:06 +01:00
1 parent fc0be86182
commit 0c6dca2fae
3 files changed
+31 -20

No files matched your search

+11 -7
View File
@@ -245,9 +245,9 @@ ggml_tensor * llama_model_granite::graph::build_layer_ffn(
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,
model.layers[il].ffn_up, model.layers[il].ffn_up_b, model.layers[il].ffn_up_s,
model.layers[il].ffn_gate, model.layers[il].ffn_gate_b, model.layers[il].ffn_gate_s,
model.layers[il].ffn_down, model.layers[il].ffn_down_b, model.layers[il].ffn_down_s,
NULL,
LLM_FFN_SILU, LLM_FFN_PAR, il);
cb(cur, "ffn_out", il);
@@ -269,15 +269,19 @@ ggml_tensor * llama_model_granite::graph::build_layer_ffn(
LLM_FFN_SILU, true,
hparams.expert_weights_scale,
LLAMA_EXPERT_GATING_FUNC_TYPE_SOFTMAX,
il);
il,
nullptr, nullptr,
model.layers[il].ffn_up_exps_s,
model.layers[il].ffn_gate_exps_s,
model.layers[il].ffn_down_exps_s);
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,
model.layers[il].ffn_up_shexp, NULL, model.layers[il].ffn_up_shexp_s,
model.layers[il].ffn_gate_shexp, NULL, model.layers[il].ffn_gate_shexp_s,
model.layers[il].ffn_down_shexp, NULL, model.layers[il].ffn_down_shexp_s,
NULL,
LLM_FFN_SILU, LLM_FFN_PAR, il);
cb(ffn_shexp, "ffn_shexp", il);