summaryrefslogtreecommitdiff
path: root/llama.cpp
diff options
context:
space:
mode:
Diffstat (limited to 'llama.cpp')
-rw-r--r--llama.cpp280
1 files changed, 0 insertions, 280 deletions
diff --git a/llama.cpp b/llama.cpp
index 2025e455..863961f1 100644
--- a/llama.cpp
+++ b/llama.cpp
@@ -202,7 +202,6 @@ enum llm_arch {
LLM_ARCH_GPTNEOX,
LLM_ARCH_MPT,
LLM_ARCH_STARCODER,
- LLM_ARCH_PERSIMMON,
LLM_ARCH_REFACT,
LLM_ARCH_BERT,
LLM_ARCH_NOMIC_BERT,
@@ -239,7 +238,6 @@ static const std::map<llm_arch, const char *> LLM_ARCH_NAMES = {
{ LLM_ARCH_MPT, "mpt" },
{ LLM_ARCH_BAICHUAN, "baichuan" },
{ LLM_ARCH_STARCODER, "starcoder" },
- { LLM_ARCH_PERSIMMON, "persimmon" },
{ LLM_ARCH_REFACT, "refact" },
{ LLM_ARCH_BERT, "bert" },
{ LLM_ARCH_NOMIC_BERT, "nomic-bert" },
@@ -596,23 +594,6 @@ static const std::map<llm_arch, std::map<llm_tensor, std::string>> LLM_TENSOR_NA
},
},
{
- LLM_ARCH_PERSIMMON,
- {
- { LLM_TENSOR_TOKEN_EMBD, "token_embd"},
- { LLM_TENSOR_OUTPUT_NORM, "output_norm"},
- { LLM_TENSOR_OUTPUT, "output"},
- { LLM_TENSOR_ATTN_NORM, "blk.%d.attn_norm"},
- { LLM_TENSOR_ATTN_QKV, "blk.%d.attn_qkv"},
- { LLM_TENSOR_ATTN_OUT, "blk.%d.attn_output"},
- { LLM_TENSOR_ATTN_Q_NORM, "blk.%d.attn_q_norm"},
- { LLM_TENSOR_ATTN_K_NORM, "blk.%d.attn_k_norm"},
- { LLM_TENSOR_FFN_NORM, "blk.%d.ffn_norm"},
- { LLM_TENSOR_FFN_DOWN, "blk.%d.ffn_down"},
- { LLM_TENSOR_FFN_UP, "blk.%d.ffn_up"},
- { LLM_TENSOR_ATTN_ROT_EMBD, "blk.%d.attn_rot_embd"},
- },
- },
- {
LLM_ARCH_MPT,
{
{ LLM_TENSOR_TOKEN_EMBD, "token_embd" },
@@ -3967,14 +3948,6 @@ static void llm_load_hparams(
default: model.type = e_model::MODEL_UNKNOWN;
}
} break;
- case LLM_ARCH_PERSIMMON:
- {
- ml.get_key(LLM_KV_ATTENTION_LAYERNORM_EPS, hparams.f_norm_eps);
- switch (hparams.n_layer) {
- case 36: model.type = e_model::MODEL_8B; break;
- default: model.type = e_model::MODEL_UNKNOWN;
- }
- } break;
case LLM_ARCH_REFACT:
{
ml.get_key(LLM_KV_ATTENTION_LAYERNORM_RMS_EPS, hparams.f_norm_rms_eps);
@@ -5221,47 +5194,6 @@ static bool llm_load_tensors(
layer.ffn_up_b = ml.create_tensor(ctx_layer, tn(LLM_TENSOR_FFN_UP, "bias", i), {n_ff});
}
} break;
- case LLM_ARCH_PERSIMMON:
- {
- model.tok_embd = ml.create_tensor(ctx_input, tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab});
-
- {
- model.output_norm = ml.create_tensor(ctx_output, tn(LLM_TENSOR_OUTPUT_NORM, "weight"), {n_embd});
- model.output_norm_b = ml.create_tensor(ctx_output, tn(LLM_TENSOR_OUTPUT_NORM, "bias"), {n_embd});
- model.output = ml.create_tensor(ctx_output_split, tn(LLM_TENSOR_OUTPUT, "weight"), {n_embd, n_vocab});
- }
-
- for (int 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.attn_norm_b = ml.create_tensor(ctx_layer, tn(LLM_TENSOR_ATTN_NORM, "bias", i), {n_embd});
-
- layer.wqkv = ml.create_tensor(ctx_split, tn(LLM_TENSOR_ATTN_QKV, "weight", i), {n_embd, n_embd + 2*n_embd_gqa});
- layer.bqkv = ml.create_tensor(ctx_layer, tn(LLM_TENSOR_ATTN_QKV, "bias", i), {n_embd + 2*n_embd_gqa});
-
- layer.wo = ml.create_tensor(ctx_split, tn(LLM_TENSOR_ATTN_OUT, "weight", i), {n_embd, n_embd});
- layer.bo = ml.create_tensor(ctx_layer, tn(LLM_TENSOR_ATTN_OUT, "bias", i), {n_embd});
-
- layer.ffn_down = ml.create_tensor(ctx_split, tn(LLM_TENSOR_FFN_DOWN, "weight", i), {n_ff, n_embd});
- layer.ffn_down_b = ml.create_tensor(ctx_layer, tn(LLM_TENSOR_FFN_DOWN, "bias", i), {n_embd});
-
- layer.ffn_up = ml.create_tensor(ctx_split, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff});
- layer.ffn_up_b = ml.create_tensor(ctx_layer, tn(LLM_TENSOR_FFN_UP, "bias", i), {n_ff});
-
- layer.ffn_norm = ml.create_tensor(ctx_layer, tn(LLM_TENSOR_FFN_NORM, "weight", i), {n_embd});
- layer.ffn_norm_b = ml.create_tensor(ctx_layer, tn(LLM_TENSOR_FFN_NORM, "bias", i), {n_embd});
-
- layer.attn_q_norm = ml.create_tensor(ctx_layer, tn(LLM_TENSOR_ATTN_Q_NORM, "weight", i), {64});
- layer.attn_q_norm_b = ml.create_tensor(ctx_layer, tn(LLM_TENSOR_ATTN_Q_NORM, "bias", i), {64});
-
- layer.attn_k_norm = ml.create_tensor(ctx_layer, tn(LLM_TENSOR_ATTN_K_NORM, "weight", i), {64});
- layer.attn_k_norm_b = ml.create_tensor(ctx_layer, tn(LLM_TENSOR_ATTN_K_NORM, "bias", i), {64});
- }
- } break;
case LLM_ARCH_BERT:
case LLM_ARCH_NOMIC_BERT:
{
@@ -7923,213 +7855,6 @@ struct llm_build_context {
return gf;
}
- struct ggml_cgraph * build_persimmon() {
- struct ggml_cgraph * gf = ggml_new_graph_custom(ctx0, LLAMA_MAX_NODES, false);
-
- 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/2 == hparams.n_rot);
-
- struct ggml_tensor * cur;
- struct ggml_tensor * inpL;
-
- inpL = llm_build_inp_embd(ctx0, lctx, hparams, batch, model.tok_embd, cb);
-
- // inp_pos - contains the positions
- struct ggml_tensor * inp_pos = build_inp_pos();
-
- // KQ_mask (mask for 1 head, it will be broadcasted to all heads)
- struct ggml_tensor * KQ_mask = build_inp_KQ_mask();
-
- for (int il = 0; il < n_layer; ++il) {
- struct ggml_tensor * residual = inpL;
-
- cur = llm_build_norm(ctx0, inpL, hparams,
- model.layers[il].attn_norm,
- model.layers[il].attn_norm_b,
- LLM_NORM, cb, il);
- cb(cur, "attn_norm", il);
-
- // self attention
- {
- cur = ggml_mul_mat(ctx0, model.layers[il].wqkv, cur);
- cb(cur, "wqkv", il);
-
- cur = ggml_add(ctx0, cur, model.layers[il].bqkv);
- cb(cur, "bqkv", il);
-
- // split qkv
- GGML_ASSERT(n_head_kv == n_head);
-
- struct ggml_tensor * tmpqkv = ggml_reshape_4d(ctx0, cur, n_embd_head, 3, n_head, n_tokens);
- cb(tmpqkv, "tmpqkv", il);
-
- struct ggml_tensor * tmpqkv_perm = ggml_cont(ctx0, ggml_permute(ctx0, tmpqkv, 0, 3, 1, 2));
- cb(tmpqkv_perm, "tmpqkv", il);
-
- struct ggml_tensor * tmpq = ggml_view_3d(
- ctx0, tmpqkv_perm, n_embd_head, n_head, n_tokens,
- ggml_element_size(tmpqkv_perm) * n_embd_head,
- ggml_element_size(tmpqkv_perm) * n_embd_head * n_head,
- 0
- );
- cb(tmpq, "tmpq", il);
-
- struct ggml_tensor * tmpk = ggml_view_3d(
- ctx0, tmpqkv_perm, n_embd_head, n_head, n_tokens,
- ggml_element_size(tmpqkv_perm) * n_embd_head,
- ggml_element_size(tmpqkv_perm) * n_embd_head * n_head,
- ggml_element_size(tmpqkv_perm) * n_embd_head * n_head * n_tokens
- );
- cb(tmpk, "tmpk", il);
-
- // Q/K Layernorm
- tmpq = llm_build_norm(ctx0, tmpq, hparams,
- model.layers[il].attn_q_norm,
- model.layers[il].attn_q_norm_b,
- LLM_NORM, cb, il);
- cb(tmpq, "tmpq", il);
-
- tmpk = llm_build_norm(ctx0, tmpk, hparams,
- model.layers[il].attn_k_norm,
- model.layers[il].attn_k_norm_b,
- LLM_NORM, cb, il);
- cb(tmpk, "tmpk", il);
-
- // RoPE the first n_rot of q/k, pass the other half, and concat.
- struct ggml_tensor * qrot = ggml_view_3d(
- ctx0, tmpq, n_rot, n_head, n_tokens,
- ggml_element_size(tmpq) * n_embd_head,
- ggml_element_size(tmpq) * n_embd_head * n_head,
- 0
- );
- cb(qrot, "qrot", il);
-
- struct ggml_tensor * krot = ggml_view_3d(
- ctx0, tmpk, n_rot, n_head, n_tokens,
- ggml_element_size(tmpk) * n_embd_head,
- ggml_element_size(tmpk) * n_embd_head * n_head,
- 0
- );
- cb(krot, "krot", il);
-
- // get the second half of tmpq, e.g tmpq[n_rot:, :, :]
- struct ggml_tensor * qpass = ggml_view_3d(
- ctx0, tmpq, n_rot, n_head, n_tokens,
- ggml_element_size(tmpq) * n_embd_head,
- ggml_element_size(tmpq) * n_embd_head * n_head,
- ggml_element_size(tmpq) * n_rot
- );
- cb(qpass, "qpass", il);
-
- struct ggml_tensor * kpass = ggml_view_3d(
- ctx0, tmpk, n_rot, n_head, n_tokens,
- ggml_element_size(tmpk) * n_embd_head,
- ggml_element_size(tmpk) * n_embd_head * n_head,
- ggml_element_size(tmpk) * n_rot
- );
- cb(kpass, "kpass", il);
-
- struct ggml_tensor * qrotated = ggml_rope_custom(
- ctx0, qrot, inp_pos, n_rot, rope_type, 0, n_orig_ctx,
- freq_base, freq_scale, ext_factor, attn_factor, beta_fast, beta_slow
- );
- cb(qrotated, "qrotated", il);
-
- struct ggml_tensor * krotated = ggml_rope_custom(
- ctx0, krot, inp_pos, n_rot, rope_type, 0, n_orig_ctx,
- freq_base, freq_scale, ext_factor, attn_factor, beta_fast, beta_slow
- );
- cb(krotated, "krotated", il);
-
- // ggml currently only supports concatenation on dim=2
- // so we need to permute qrot, qpass, concat, then permute back.
- qrotated = ggml_cont(ctx0, ggml_permute(ctx0, qrotated, 2, 1, 0, 3));
- cb(qrotated, "qrotated", il);
-
- krotated = ggml_cont(ctx0, ggml_permute(ctx0, krotated, 2, 1, 0, 3));
- cb(krotated, "krotated", il);
-
- qpass = ggml_cont(ctx0, ggml_permute(ctx0, qpass, 2, 1, 0, 3));
- cb(qpass, "qpass", il);
-
- kpass = ggml_cont(ctx0, ggml_permute(ctx0, kpass, 2, 1, 0, 3));
- cb(kpass, "kpass", il);
-
- struct ggml_tensor * Qcur = ggml_concat(ctx0, qrotated, qpass);
- cb(Qcur, "Qcur", il);
-
- struct ggml_tensor * Kcur = ggml_concat(ctx0, krotated, kpass);
- cb(Kcur, "Kcur", il);
-
- struct ggml_tensor * Q = ggml_cont(ctx0, ggml_permute(ctx0, Qcur, 2, 1, 0, 3));
- cb(Q, "Q", il);
-
- Kcur = ggml_cont(ctx0, ggml_permute(ctx0, Kcur, 2, 1, 0, 3));
- cb(Kcur, "Kcur", il);
-
- struct ggml_tensor * Vcur = ggml_view_3d(
- ctx0, tmpqkv_perm, n_embd_head, n_head, n_tokens,
- ggml_element_size(tmpqkv_perm) * n_embd_head,
- ggml_element_size(tmpqkv_perm) * n_embd_head * n_head,
- ggml_element_size(tmpqkv_perm) * n_embd_head * n_head * n_tokens * 2
- );
- cb(Vcur, "Vcur", il);
-
- cur = llm_build_kv(ctx0, model, hparams, cparams, kv_self, gf,
- model.layers[il].wo, model.layers[il].bo,
- Kcur, Vcur, Q, KQ_mask, n_tokens, kv_head, n_kv, 1.0f/sqrtf(float(n_embd_head)), cb, il);
- }
-
- if (il == n_layer - 1) {
- // skip computing output for unused tokens
- struct ggml_tensor * inp_out_ids = build_inp_out_ids();
- cur = ggml_get_rows(ctx0, cur, inp_out_ids);
- residual = ggml_get_rows(ctx0, residual, inp_out_ids);
- }
-
- struct ggml_tensor * ffn_inp = ggml_add(ctx0, residual, cur);
- cb(ffn_inp, "ffn_inp", il);
-
- // feed-forward network
- {
- cur = llm_build_norm(ctx0, ffn_inp, hparams,
- model.layers[il].ffn_norm,
- model.layers[il].ffn_norm_b,
- LLM_NORM, cb, il);
- cb(cur, "ffn_norm", il);
-
- cur = llm_build_ffn(ctx0, cur,
- model.layers[il].ffn_up, model.layers[il].ffn_up_b,
- NULL, NULL,
- model.layers[il].ffn_down, model.layers[il].ffn_down_b,
- NULL,
- LLM_FFN_RELU_SQR, LLM_FFN_SEQ, cb, il);
- cb(cur, "ffn_out", il);
- }
-
- cur = ggml_add(ctx0, cur, ffn_inp);
- cb(cur, "l_out", il);
-
- inpL = cur;
- }
-
- cur = inpL;
-
- cur = llm_build_norm(ctx0, cur, hparams,
- model.output_norm,
- model.output_norm_b,
- LLM_NORM, cb, -1);
- cb(cur, "result_norm", -1);
-
- cur = ggml_mul_mat(ctx0, model.output, cur);
- cb(cur, "result_output", -1);
-
- ggml_build_forward_expand(gf, cur);
-
- return gf;
- }
-
struct ggml_cgraph * build_refact() {
struct ggml_cgraph * gf = ggml_new_graph_custom(ctx0, LLAMA_MAX_NODES, false);
@@ -10898,10 +10623,6 @@ static struct ggml_cgraph * llama_build_graph(
{
result = llm.build_starcoder();
} break;
- case LLM_ARCH_PERSIMMON:
- {
- result = llm.build_persimmon();
- } break;
case LLM_ARCH_REFACT:
{
result = llm.build_refact();
@@ -15992,7 +15713,6 @@ enum llama_rope_type llama_rope_type(const struct llama_model * model) {
case LLM_ARCH_FALCON:
case LLM_ARCH_GROK:
case LLM_ARCH_DBRX:
- case LLM_ARCH_PERSIMMON:
case LLM_ARCH_BERT:
case LLM_ARCH_NOMIC_BERT:
case LLM_ARCH_STABLELM:
generated by cgit v1.2.3 (git 2.25.1) at 2025-08-17 22:27:43 +0000