Open
Description
Name and Version
% llama-cli --version
load_backend: loaded RPC backend from /home/bsahd/llama.cpp-prebuilt/build/bin/libggml-rpc.so
load_backend: loaded CPU backend from /home/bsahd/llama.cpp-prebuilt/build/bin/libggml-cpu-haswell.so
version: 5815 (d4cdd9c)
built with cc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0 for x86_64-linux-gnu
I'm using llama-b5815-bin-ubuntu-x64.zip
Operating systems
Linux
GGML backends
CPU
Hardware
AMD Ryzen 5 3500U
Models
https://huggingface.co/zhiyuan8/RWKV-v7-0.1B-G1-GGUF/blob/main/rwkv7-0.1B-g1-F16.gguf
Problem description & steps to reproduce
Crashes with "zsh: IOT instruction (core dumped)" when using reverse prompt.
Crash occurs regardless of whether --no-mmap is used or not.
Also occurred at https://huggingface.co/zhiyuan8/RWKV-v7-1.5B-World-v3-GGUF/blob/main/rwkv7-1.5B-world-Q8_0.gguf.
First Bad Commit
No response
Relevant log output
% llama-cli --no-mmap -m rwkv7-0.1B-g1-F16.gguf -r "User:"
load_backend: loaded RPC backend from /home/bsahd/llama.cpp-prebuilt/build/bin/libggml-rpc.so
load_backend: loaded CPU backend from /home/bsahd/llama.cpp-prebuilt/build/bin/libggml-cpu-haswell.so
build: 5815 (d4cdd9c1) with cc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0 for x86_64-linux-gnu
main: llama backend init
main: load the model and apply lora adapter, if any
llama_model_loader: loaded meta data with 33 key-value pairs and 342 tensors from rwkv7-0.1B-g1-F16.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 = rwkv7
llama_model_loader: - kv 1: general.type str = model
llama_model_loader: - kv 2: general.name str = Rwkv7 0.1B G1
llama_model_loader: - kv 3: general.finetune str = g1
llama_model_loader: - kv 4: general.basename str = rwkv7
llama_model_loader: - kv 5: general.size_label str = 0.1B
llama_model_loader: - kv 6: general.license str = apache-2.0
llama_model_loader: - kv 7: general.base_model.count u32 = 1
llama_model_loader: - kv 8: general.base_model.0.name str = Rwkv7 G1
llama_model_loader: - kv 9: general.base_model.0.organization str = BlinkDL
llama_model_loader: - kv 10: general.base_model.0.repo_url str = https://huggingface.co/BlinkDL/rwkv7-g1
llama_model_loader: - kv 11: general.tags arr[str,1] = ["text-generation"]
llama_model_loader: - kv 12: general.languages arr[str,8] = ["en", "zh", "ja", "ko", "fr", "ar", ...
llama_model_loader: - kv 13: rwkv7.context_length u32 = 1048576
llama_model_loader: - kv 14: rwkv7.embedding_length u32 = 768
llama_model_loader: - kv 15: rwkv7.block_count u32 = 12
llama_model_loader: - kv 16: rwkv7.attention.layer_norm_epsilon f32 = 0.000010
llama_model_loader: - kv 17: rwkv7.wkv.head_size u32 = 64
llama_model_loader: - kv 18: rwkv7.attention.decay_lora_rank u32 = 64
llama_model_loader: - kv 19: rwkv7.attention.iclr_lora_rank u32 = 64
llama_model_loader: - kv 20: rwkv7.attention.value_residual_mix_lora_rank u32 = 32
llama_model_loader: - kv 21: rwkv7.attention.gate_lora_rank u32 = 128
llama_model_loader: - kv 22: rwkv7.feed_forward_length u32 = 3072
llama_model_loader: - kv 23: general.file_type u32 = 1
llama_model_loader: - kv 24: rwkv7.attention.head_count u32 = 0
llama_model_loader: - kv 25: tokenizer.ggml.model str = rwkv
llama_model_loader: - kv 26: tokenizer.ggml.tokens arr[str,65536] = ["<s>", "\\x00", "\\x01", "\\x02", "\...
llama_model_loader: - kv 27: tokenizer.ggml.token_type arr[i32,65536] = [3, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv 28: tokenizer.ggml.bos_token_id u32 = 0
llama_model_loader: - kv 29: tokenizer.ggml.eos_token_id u32 = 0
llama_model_loader: - kv 30: tokenizer.ggml.eot_token_id u32 = 261
llama_model_loader: - kv 31: tokenizer.chat_template str = rwkv-world
llama_model_loader: - kv 32: general.quantization_version u32 = 2
llama_model_loader: - type f32: 196 tensors
llama_model_loader: - type f16: 146 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type = F16
print_info: file size = 367.15 MiB (16.11 BPW)
load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
load: special_eot_id is not in special_eog_ids - the tokenizer config may be incorrect
load: special tokens cache size = 1
load: token to piece cache size = 0.3561 MB
print_info: arch = rwkv7
print_info: vocab_only = 0
print_info: n_ctx_train = 1048576
print_info: n_embd = 768
print_info: n_layer = 12
print_info: n_head = 0
print_info: n_head_kv = 0
print_info: n_rot = 0
print_info: n_swa = 0
print_info: is_swa_any = 0
print_info: n_embd_head_k = 0
print_info: n_embd_head_v = 0
print_info: n_gqa = 0
print_info: n_embd_k_gqa = 0
print_info: n_embd_v_gqa = 0
print_info: f_norm_eps = 1.0e-05
print_info: f_norm_rms_eps = 0.0e+00
print_info: f_clamp_kqv = 0.0e+00
print_info: f_max_alibi_bias = 0.0e+00
print_info: f_logit_scale = 0.0e+00
print_info: f_attn_scale = 0.0e+00
print_info: n_ff = 3072
print_info: n_expert = 0
print_info: n_expert_used = 0
print_info: causal attn = 1
print_info: pooling type = 0
print_info: rope type = -1
print_info: rope scaling = linear
print_info: freq_base_train = 10000.0
print_info: freq_scale_train = 1
print_info: n_ctx_orig_yarn = 1048576
print_info: rope_finetuned = unknown
print_info: model type = 190M
print_info: model params = 191.13 M
print_info: general.name = Rwkv7 0.1B G1
print_info: vocab type = RWKV
print_info: n_vocab = 65536
print_info: n_merges = 0
print_info: BOS token = 0 '<s>'
print_info: EOS token = 0 '<s>'
print_info: EOT token = 261 '\n\n'
print_info: LF token = 11 '\n'
print_info: EOG token = 0 '<s>'
print_info: EOG token = 261 '\n\n'
print_info: max token length = 192
load_tensors: loading model tensors, this can take a while... (mmap = false)
load_tensors: offloading 0 repeating layers to GPU
load_tensors: offloaded 0/13 layers to GPU
load_tensors: CPU model buffer size = 367.15 MiB
......................................
llama_context: constructing llama_context
llama_context: n_seq_max = 1
llama_context: n_ctx = 4096
llama_context: n_ctx_per_seq = 4096
llama_context: n_batch = 2048
llama_context: n_ubatch = 512
llama_context: causal_attn = 1
llama_context: flash_attn = 0
llama_context: freq_base = 10000.0
llama_context: freq_scale = 1
llama_context: n_ctx_per_seq (4096) < n_ctx_train (1048576) -- the full capacity of the model will not be utilized
llama_context: CPU output buffer size = 0.25 MiB
llama_memory_recurrent: mem_size = 1, n_seq_max = 1, type_r = 'f32', type_s = 'f32', n_layer = 12
llama_memory_recurrent: CPU KV buffer size = 2.32 MiB
llama_memory_recurrent: KV self size = 2.32 MiB, R (f32): 0.07 MiB, S (f32): 2.25 MiB
llama_context: CPU compute buffer size = 129.50 MiB
llama_context: graph nodes = 1420
llama_context: graph splits = 1
common_init_from_params: setting dry_penalty_last_n to ctx_size = 4096
common_init_from_params: warming up the model with an empty run - please wait ... (--no-warmup to disable)
Failed to infer a tool call example (possible template bug)
main: llama threadpool init, n_threads = 4
main: chat template example:
System: You are a helpful assistant
User: Hello
Assistant: Hi there
User: How are you?
Assistant:
system_info: n_threads = 4 (n_threads_batch = 4) / 8 | CPU : SSE3 = 1 | SSSE3 = 1 | AVX = 1 | AVX2 = 1 | F16C = 1 | FMA = 1 | BMI2 = 1 | LLAMAFILE = 1 | OPENMP = 1 | REPACK = 1 |
main: interactive mode on.
Reverse prompt: 'User:'
sampler seed: 1308970135
sampler params:
repeat_last_n = 64, repeat_penalty = 1.000, frequency_penalty = 0.000, presence_penalty = 0.000
dry_multiplier = 0.000, dry_base = 1.750, dry_allowed_length = 2, dry_penalty_last_n = 4096
top_k = 40, top_p = 0.950, min_p = 0.050, xtc_probability = 0.000, xtc_threshold = 0.100, typical_p = 1.000, top_n_sigma = -1.000, temp = 0.800
mirostat = 0, mirostat_lr = 0.100, mirostat_ent = 5.000
sampler chain: logits -> logit-bias -> penalties -> dry -> top-n-sigma -> top-k -> typical -> top-p -> min-p -> xtc -> temp-ext -> dist
generate: n_ctx = 4096, n_batch = 2048, n_predict = -1, n_keep = 0
== Running in interactive mode. ==
- Press Ctrl+C to interject at any time.
- Press Return to return control to the AI.
- To return control without starting a new line, end your input with '/'.
- If you want to submit another line, end your input with '\'.
- Not using system message. To change it, set a different value via -sys PROMPT
[New LWP 8041]
[New LWP 8040]
[New LWP 8039]
[New LWP 8035]
This GDB supports auto-downloading debuginfo from the following URLs:
<https://debuginfod.archlinux.org>
Enable debuginfod for this session? (y or [n]) [answered N; input not from terminal]
Debuginfod has been disabled.
To make this setting permanent, add 'set debuginfod enabled off' to .gdbinit.
Function(s) ^std::(move|forward|as_const|(__)?addressof) will be skipped when stepping.
Function(s) ^std::(shared|unique)_ptr<.*>::(get|operator) will be skipped when stepping.
Function(s) ^std::(basic_string|vector|array|deque|(forward_)?list|(unordered_|flat_)?(multi)?(map|set)|span)<.*>::(c?r?(begin|end)|front|back|data|size|empty) will be skipped when stepping.
Function(s) ^std::(basic_string|vector|array|deque|span)<.*>::operator.] will be skipped when stepping.
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/usr/lib/libthread_db.so.1".
0x00007f4102f88e22 in ?? () from /usr/lib/libc.so.6
#0 0x00007f4102f88e22 in ?? () from /usr/lib/libc.so.6
#1 0x00007f4102f7cfda in ?? () from /usr/lib/libc.so.6
#2 0x00007f4102f7d024 in ?? () from /usr/lib/libc.so.6
#3 0x00007f4102fed92f in wait4 () from /usr/lib/libc.so.6
#4 0x00007f41035db16b in ggml_print_backtrace () from /home/bsahd/llama.cpp-prebuilt/build/bin/libggml-base.so
#5 0x00007f41035ebc8f in ggml_uncaught_exception() () from /home/bsahd/llama.cpp-prebuilt/build/bin/libggml-base.so
#6 0x00007f41032b1c1a in __cxxabiv1::__terminate (handler=<optimized out>) at /usr/src/debug/gcc/gcc/libstdc++-v3/libsupc++/eh_terminate.cc:48
warning: 48 /usr/src/debug/gcc/gcc/libstdc++-v3/libsupc++/eh_terminate.cc: そのようなファイルやディレクトリはありません
#7 0x00007f41032975db in std::terminate () at /usr/src/debug/gcc/gcc/libstdc++-v3/libsupc++/eh_terminate.cc:58
58 in /usr/src/debug/gcc/gcc/libstdc++-v3/libsupc++/eh_terminate.cc
#8 0x00007f41032b1ed6 in __cxxabiv1::__cxa_throw (obj=<optimized out>, tinfo=0x7f4103484d90 <typeinfo for std::runtime_error>, dest=0x7f41032c99b0 <std::runtime_error::~runtime_error()>) at /usr/src/debug/gcc/gcc/libstdc++-v3/libsupc++/eh_throw.cc:98
warning: 98 /usr/src/debug/gcc/gcc/libstdc++-v3/libsupc++/eh_throw.cc: そのようなファイルやディレクトリはありません
#9 0x000055d87117e258 in common_sampler_last(common_sampler const*) [clone .cold] ()
#10 0x000055d8711878bd in main ()
[Inferior 1 (process 8034) detached]
terminate called after throwing an instance of 'std::runtime_error'
what(): ring buffer: index out of bounds
zsh: IOT instruction (core dumped) llama-cli --no-mmap -cnv -m rwkv7-0.1B-g1-F16.gguf -r "User:"