vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

add extra tensors

Browse source
This commit is contained in:
akawrykow 2023-08-29 18:28:04 -07:00
parent 66a6dbf702
commit 999f1f879d
1 changed files with 33 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

37
llama.cpp
View file

@ -321,6 +321,7 @@ static std::map<llm_arch, std::map<llm_tensor, std::string>> LLM_TENSOR_NAMES =
{ LLM_TENSOR_ATTN_NORM_2, "blk.%d.attn_norm_2" }, { LLM_TENSOR_ATTN_NORM_2, "blk.%d.attn_norm_2" },
{ LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" }, { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv" },
{ LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" }, { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output" },
{ LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm" },
{ LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" }, { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down" },
{ LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" }, { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up" },
}, },
@ -900,15 +901,20 @@ struct llama_layer {
struct ggml_tensor * wk; struct ggml_tensor * wk;
struct ggml_tensor * wv; struct ggml_tensor * wv;
struct ggml_tensor * wo; struct ggml_tensor * wo;
struct ggml_tensor * wo_b;
struct ggml_tensor * wqkv; struct ggml_tensor * wqkv;
struct ggml_tensor * wqkv_b;
// normalization // normalization
struct ggml_tensor * ffn_norm; struct ggml_tensor * ffn_norm;
struct ggml_tensor * ffn_norm_b;
// ff // ff
struct ggml_tensor * w1; // ffn_gate struct ggml_tensor * w1; // ffn_gate
struct ggml_tensor * w2; // ffn_down struct ggml_tensor * w2; // ffn_down
struct ggml_tensor * w3; // ffn_up struct ggml_tensor * w2_b; // ff_down bias
struct ggml_tensor * w3; // ffn_up
struct ggml_tensor * w3_b; // ff_up bias
}; };
struct llama_kv_cache { struct llama_kv_cache {
@ -2012,14 +2018,37 @@ static void llm_load_tensors(
// I think this is because we skip the QKV reshaping in the conversion script (maybe because parallel attention is disabled?) // I think this is because we skip the QKV reshaping in the conversion script (maybe because parallel attention is disabled?)
if (model.type == MODEL_1B) { if (model.type == MODEL_1B) {
layer.wqkv = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_QKV, "weight", i), {n_embd, n_embd * 3}, backend_split); layer.wqkv = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_QKV, "weight", i), {n_embd, n_embd * 3}, backend_split);
// TODO - The config.json has a `bias: true` -- can we plumb that through here to figure out if we need to include it?
layer.wqkv_b = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_QKV, "bias", i), {n_embd * 3}, backend);
} else { } else {
layer.wqkv = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_QKV, "weight", i), {n_embd, n_embd + 2*n_embd_gqa}, backend_split); layer.wqkv = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_QKV, "weight", i), {n_embd, n_embd + 2*n_embd_gqa}, backend_split);
} }
layer.wo = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_OUT, "weight", i), {n_embd, n_embd}, backend_split); layer.wo = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_OUT, "weight", i), {n_embd, n_embd}, backend_split);
// TODO - The config.json has a `bias: true` -- can we plumb that through here to figure out if we need to include it?
if (model.type == MODEL_1B) {
layer.wo_b = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_OUT, "bias", i), {n_embd}, backend_split);
}
// TODO: Can we figure out if we need this dynamically?
if (model.type == MODEL_1B) {
layer.ffn_norm = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_NORM, "weight", i), {n_embd}, backend);
layer.ffn_norm_b = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_NORM, "bias", i), {n_embd}, backend);
}
layer.w2 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "weight", i), {n_ff, n_embd}, backend_split);
// TODO - The config.json has a `bias: true` -- can we plumb that through here to figure out if we need to include it?
if (model.type == MODEL_1B) {
layer.w2_b = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "bias", i), {n_embd}, backend_split);
}
layer.w2 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "weight", i), { n_ff, n_embd}, backend_split);
layer.w3 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff}, backend_split); layer.w3 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff}, backend_split);
// TODO - The config.json has a `bias: true` -- can we plumb that through here to figure out if we need to include it?
if (model.type == MODEL_1B) {
// TODO - where does 4 come from?
layer.w3_b = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "bias", i), {n_embd * 4}, backend_split);
}
if (backend == GGML_BACKEND_GPU) { if (backend == GGML_BACKEND_GPU) {
vram_weights += vram_weights +=