rnwang04
commented
7 months ago Hi @ktjylsj , we have reproduced your error.
This error is triggered by llm_load_tensors: offloaded 33/49 layers to GPU:
- Solar 10.7b has 49 model layers, and you just transfer 33 layers to GPU by setting
-ngl 33 - our latest code (ipex-llm[cpp] > 2.5.0b20240401) has bug with this case, we will try to fix it
But as for now, one simple way to solve this error is just set -ngl 49 or bigger numbers 😊
ktjylsj
Dear IPEX team.
GGML_ASSERT occurs when running the Solar 10.7b model in llama-cpp. Here's the output when running the model. I've tested the below models by llama-cpp and same GGML_ASSERT occured. https://huggingface.co/TheBloke/SOLAR-10.7B-Instruct-v1.0-GGUF/blob/main/solar-10.7b-instruct-v1.0.Q6_K.gguf https://huggingface.co/upstage/SOLAR-10.7B-Instruct-v1.0
", "", "<0x00>", "<... llama_model_loader: - kv 14: tokenizer.ggml.scores arr[f32,32000] = [0.000000, 0.000000, 0.000000, 0.0000... llama_model_loader: - kv 15: tokenizer.ggml.token_type arr[i32,32000] = [2, 3, 3, 6, 6, 6, 6, 6, 6, 6, 6, 6, ... llama_model_loader: - kv 16: tokenizer.ggml.bos_token_id u32 = 1 llama_model_loader: - kv 17: tokenizer.ggml.eos_token_id u32 = 2 llama_model_loader: - kv 18: tokenizer.ggml.unknown_token_id u32 = 0 llama_model_loader: - kv 19: tokenizer.ggml.padding_token_id u32 = 2 llama_model_loader: - kv 20: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 21: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 22: tokenizer.chat_template str = {% for message in messages %}{% if me... llama_model_loader: - kv 23: general.quantization_version u32 = 2 llama_model_loader: - type f32: 97 tensors llama_model_loader: - type q6_K: 338 tensors llm_load_vocab: special tokens definition check successful ( 259/32000 ). llm_load_print_meta: format = GGUF V3 (latest) llm_load_print_meta: arch = llama llm_load_print_meta: vocab type = SPM llm_load_print_meta: n_vocab = 32000 llm_load_print_meta: n_merges = 0 llm_load_print_meta: n_ctx_train = 16384 llm_load_print_meta: n_embd = 4096 llm_load_print_meta: n_head = 32 llm_load_print_meta: n_head_kv = 8 llm_load_print_meta: n_layer = 48 llm_load_print_meta: n_rot = 128 llm_load_print_meta: n_embd_head_k = 128 llm_load_print_meta: n_embd_head_v = 128 llm_load_print_meta: n_gqa = 4 llm_load_print_meta: n_embd_k_gqa = 1024 llm_load_print_meta: n_embd_v_gqa = 1024 llm_load_print_meta: f_norm_eps = 0.0e+00 llm_load_print_meta: f_norm_rms_eps = 1.0e-05 llm_load_print_meta: f_clamp_kqv = 0.0e+00 llm_load_print_meta: f_max_alibi_bias = 0.0e+00 llm_load_print_meta: f_logit_scale = 0.0e+00 llm_load_print_meta: n_ff = 14336 llm_load_print_meta: n_expert = 0 llm_load_print_meta: n_expert_used = 0 llm_load_print_meta: causal attn = 1 llm_load_print_meta: pooling type = 0 llm_load_print_meta: rope type = 0 llm_load_print_meta: rope scaling = linear llm_load_print_meta: freq_base_train = 10000.0 llm_load_print_meta: freq_scale_train = 1 llm_load_print_meta: n_yarn_orig_ctx = 16384 llm_load_print_meta: rope_finetuned = unknown llm_load_print_meta: ssm_d_conv = 0 llm_load_print_meta: ssm_d_inner = 0 llm_load_print_meta: ssm_d_state = 0 llm_load_print_meta: ssm_dt_rank = 0 llm_load_print_meta: model type = 34B llm_load_print_meta: model ftype = Q6_K llm_load_print_meta: model params = 10.73 B llm_load_print_meta: model size = 8.20 GiB (6.56 BPW) llm_load_print_meta: general.name = models llm_load_print_meta: BOS token = 1 '' llm_load_print_meta: EOS token = 2 '' llm_load_print_meta: UNK token = 0 'ggml_backend_sycl_set_mul_device_mode: true detect 1 SYCL GPUs: [0] with top Max compute units:512 llm_load_tensors: ggml ctx size = 0.33 MiB llm_load_tensors: offloading 33 repeating layers to GPU llm_load_tensors: offloaded 33/49 layers to GPU llm_load_tensors: SYCL0 buffer size = 5631.66 MiB llm_load_tensors: CPU buffer size = 8396.59 MiB .................................................................................................... llama_new_context_with_model: n_ctx = 512 llama_new_context_with_model: n_batch = 512 llama_new_context_with_model: n_ubatch = 512 llama_new_context_with_model: freq_base = 10000.0 llama_new_context_with_model: freq_scale = 1 llama_kv_cache_init: SYCL0 KV buffer size = 66.00 MiB llama_kv_cache_init: SYCL_Host KV buffer size = 30.00 MiB llama_new_context_with_model: KV self size = 96.00 MiB, K (f16): 48.00 MiB, V (f16): 48.00 MiB llama_new_context_with_model: SYCL_Host output buffer size = 0.12 MiB llama_new_context_with_model: SYCL0 compute buffer size = 173.04 MiB llama_new_context_with_model: SYCL_Host compute buffer size = 9.01 MiB llama_new_context_with_model: graph nodes = 1590 llama_new_context_with_model: graph splits = 169
system_info: n_threads = 8 / 20 | AVX = 1 | AVX_VNNI = 0 | AVX2 = 1 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | FMA = 1 | NEON = 0 | ARM_FMA = 0 | F16C = 1 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 1 | SSE3 = 1 | SSSE3 = 1 | VSX = 0 | MATMUL_INT8 = 0 | sampling: repeat_last_n = 64, repeat_penalty = 1.000, frequency_penalty = 0.000, presence_penalty = 0.000 top_k = 40, tfs_z = 1.000, top_p = 0.950, min_p = 0.050, typical_p = 1.000, temp = 0.800 mirostat = 0, mirostat_lr = 0.100, mirostat_ent = 5.000 sampling order: CFG -> Penalties -> top_k -> tfs_z -> typical_p -> top_p -> min_p -> temperature generate: n_ctx = 512, n_batch = 2048, n_predict = 32, n_keep = 1
Once upon a time, there existed a little girl who liked to have adventures. She wanted to go to places and meet new people, and have fun.GGML_ASSERT: C:\Users\Administrator\actions-runner\bigdl-core-cpp-release_work\llm.cpp\llm.cpp\llama-cpp-bigdl\ggml.c:14154: ne1 == N