Segmentation fault (core dumped) error

[ee2110]

Hi, I was able to use llama-cpp-python steadily last week, however, I encountered this segmentation fault error since few days ago. I have tried reinstalled again and also adjusting the n_gpu_layers value, but I still encounter the same issue. I have no idea what is the root cause of this.
Does anyone could kindly help to advice how to solve this? Thank you.

Segmentation fault (core dumped)

full log output as below:

llama_model_loader: loaded meta data with 21 key-value pairs and 363 tensors from /home/yeokeat/goal-inferences-project/llama_cpp/vicuna-13b-v1.5-16bit.gguf (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv   0:                       general.architecture str              = llama
llama_model_loader: - kv   1:                               general.name str              = .
llama_model_loader: - kv   2:                           llama.vocab_size u32              = 32000
llama_model_loader: - kv   3:                       llama.context_length u32              = 2048
llama_model_loader: - kv   4:                     llama.embedding_length u32              = 5120
llama_model_loader: - kv   5:                          llama.block_count u32              = 40
llama_model_loader: - kv   6:                  llama.feed_forward_length u32              = 13824
llama_model_loader: - kv   7:                 llama.rope.dimension_count u32              = 128
llama_model_loader: - kv   8:                 llama.attention.head_count u32              = 40
llama_model_loader: - kv   9:              llama.attention.head_count_kv u32              = 40
llama_model_loader: - kv  10:     llama.attention.layer_norm_rms_epsilon f32              = 0.000001
llama_model_loader: - kv  11:                          general.file_type u32              = 1
llama_model_loader: - kv  12:                       tokenizer.ggml.model str              = llama
llama_model_loader: - kv  13:                      tokenizer.ggml.tokens arr[str,32000]   = ["<unk>", "<s>", "</s>", "<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.padding_token_id u32              = 0
llama_model_loader: - kv  19:               tokenizer.ggml.add_bos_token bool             = true
llama_model_loader: - kv  20:               tokenizer.ggml.add_eos_token bool             = false
llama_model_loader: - type  f32:   81 tensors
llama_model_loader: - type  f16:  282 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      = 2048
llm_load_print_meta: n_embd           = 5120
llm_load_print_meta: n_head           = 40
llm_load_print_meta: n_head_kv        = 40
llm_load_print_meta: n_layer          = 40
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            = 1
llm_load_print_meta: n_embd_k_gqa     = 5120
llm_load_print_meta: n_embd_v_gqa     = 5120
llm_load_print_meta: f_norm_eps       = 0.0e+00
llm_load_print_meta: f_norm_rms_eps   = 1.0e-06
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             = 13824
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  = 2048
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       = 13B
llm_load_print_meta: model ftype      = F16
llm_load_print_meta: model params     = 13.02 B
llm_load_print_meta: model size       = 24.24 GiB (16.00 BPW) 
llm_load_print_meta: general.name     = .
llm_load_print_meta: BOS token        = 1 '<s>'
llm_load_print_meta: EOS token        = 2 '</s>'
llm_load_print_meta: UNK token        = 0 '<unk>'
llm_load_print_meta: PAD token        = 0 '<unk>'
llm_load_print_meta: LF token         = 13 '<0x0A>'
ggml_cuda_init: GGML_CUDA_FORCE_MMQ:   no
ggml_cuda_init: CUDA_USE_TENSOR_CORES: yes
ggml_cuda_init: found 2 CUDA devices:
  Device 0: NVIDIA A40, compute capability 8.6, VMM: yes
  Device 1: NVIDIA A40, compute capability 8.6, VMM: yes
llm_load_tensors: ggml ctx size =    0.42 MiB
llm_load_tensors: offloading 10 repeating layers to GPU
llm_load_tensors: offloaded 10/41 layers to GPU
llm_load_tensors:        CPU buffer size = 24826.58 MiB
llm_load_tensors:      CUDA0 buffer size =  3025.20 MiB
llm_load_tensors:      CUDA1 buffer size =  3025.20 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:  CUDA_Host KV buffer size =   300.00 MiB
llama_kv_cache_init:      CUDA0 KV buffer size =    50.00 MiB
llama_kv_cache_init:      CUDA1 KV buffer size =    50.00 MiB
llama_new_context_with_model: KV self size  =  400.00 MiB, K (f16):  200.00 MiB, V (f16):  200.00 MiB
llama_new_context_with_model:  CUDA_Host  output buffer size =     0.12 MiB
llama_new_context_with_model:      CUDA0 compute buffer size =   385.00 MiB
llama_new_context_with_model:      CUDA1 compute buffer size =    85.00 MiB
llama_new_context_with_model:  CUDA_Host compute buffer size =    21.01 MiB
llama_new_context_with_model: graph nodes  = 1286
llama_new_context_with_model: graph splits = 335
AVX = 1 | AVX_VNNI = 0 | AVX2 = 1 | AVX512 = 1 | AVX512_VBMI = 1 | AVX512_VNNI = 1 | 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 | 
Model metadata: {'tokenizer.ggml.add_eos_token': 'false', 'tokenizer.ggml.padding_token_id': '0', 'tokenizer.ggml.eos_token_id': '2', 'general.architecture': 'llama', 'llama.context_length': '2048', 'general.name': '.', 'llama.vocab_size': '32000', 'general.file_type': '1', 'tokenizer.ggml.add_bos_token': 'true', 'llama.embedding_length': '5120', 'llama.feed_forward_length': '13824', 'llama.attention.layer_norm_rms_epsilon': '0.000001', 'llama.rope.dimension_count': '128', 'tokenizer.ggml.bos_token_id': '1', 'llama.attention.head_count': '40', 'llama.block_count': '40', 'llama.attention.head_count_kv': '40', 'tokenizer.ggml.model': 'llama'}
Using fallback chat format: None
Segmentation fault (core dumped)

How I install llama-cpp-python:

export LLAMA_CUBLAS=1
CMAKE_ARGS="-DLLAMA_CUBLAS=on" FORCE_CMAKE=1 pip install llama-cpp-python

To reproduce using the same program code:

from llama_cpp import Llama
import json
import os
import sys

MODEL_PATH = '/path/to/llama_cpp/vicuna-13b-v1.5-16bit.gguf'
llm = Llama(model_path=MODEL_PATH, n_gpu_layers=10)

STRING_PROMPT = """
### System: You are an AI assistant that follows instructions exceptionally well. Be as helpful as possible.
### User: Briefly list down the steps to perform Cook Bacon.
### Assistant:
"""

def llm_infer(llm, STRING_PROMPT):
    return llm(
        prompt=STRING_PROMPT,
        max_tokens=2048,
        echo=False,
        seed=1024,
        temperature=0.001
    )


result = llm_infer(llm, STRING_PROMPT )

print(result['choices'][0]['text'])

[ismail-yussuf]

ive been getting this issue as well. not sure why

[ee2110]

Hi @ismail-yussuf , I found here the solution we can try to fix that.

setting logits_all=True at Llama(model_path=MODEL_PATH, n_gpu_layers=-1, logits_all=True)