GGML_ASSERT(seq_id < n_tokens && "seq_id cannot be larger than n_tokens with pooling_type == CLS") failed (also MEAN)

[Morphus1]

not sure if I should submit an issue about this:

getting GGML_ASSERT(seq_id < n_tokens && "seq_id cannot be larger than n_tokens with pooling_type == CLS") failed and GGML_ASSERT(seq_id < n_tokens && "seq_id cannot be larger than n_tokens with pooling_type == MEAN") failed after introducing Pooling.

I'm using the java wrapper tho, https://github.com/kherud/java-llama.cpp

only reason I think the issue is in llama.cpp is Ollama user mentioned the same issue here: ollama/ollama#4545

[aviallon]

@slaren Using build 'b5404', I am encountering the same issue with:

[user@system]$ export LLAMA_ARG_HF_REPO=nomic-ai/nomic-embed-text-v2-moe-GGUF:Q4_K_M \
LLAMA_ARG_EMBEDDINGS=1 \
LLAMA_ARG_ENDPOINT_METRICS=1 \
LLAMA_ARG_NO_WEBUI=1 \
LLAMA_ARG_HOST=0.0.0.0 \
LLAMA_ARG_N_PARALLEL=4 \
LLAMA_ARG_ALIAS=embeddings-multilingual \
LLAMA_ARG_PORT=80 \
LLAMA_ARG_CACHE_TYPE_K=f16 \
LLAMA_ARG_FLASH_ATTN=0 \
LLAMA_ARG_CTX_SIZE=2048 \
LLAMA_ARG_BATCH=448 \
LLAMA_ARG_BATCH=512 \
LLAMA_ARG_THREADS=1 \
LLAMA_ARG_N_PREDICT=-1 \
LLAMA_ARG_N_GPU_LAYERS=0 \
LLAMA_ARG_NUMA=distribute \
LLAMA_ARG_MLOCK=0 \
LLAMA_ARG_ENDPOINT_SLOTS=1 \
LLAMA_ARG_NO_CONTEXT_SHIFT=0 \
LLAMA_ARG_UBATCH=512
[user@system]$ llama-server --seed 0 --temp 0.0

Full logs

load_backend: loaded CPU backend from /app/libggml-cpu-haswell.so
warning: no usable GPU found, --gpu-layers option will be ignored
warning: one possible reason is that llama.cpp was compiled without GPU support
warning: consult docs/build.md for compilation instructions
curl_perform_with_retry: HEAD https://huggingface.co/nomic-ai/nomic-embed-text-v2-moe-GGUF/resolve/main/nomic-embed-text-v2-moe.Q4_K_M.gguf (attempt 1 of 1)...
common_download_file_single: using cached file: /root/.cache/llama.cpp/nomic-ai_nomic-embed-text-v2-moe-GGUF_nomic-embed-text-v2-moe.Q4_K_M.gguf
build: 1 (faa0b9ba) with cc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0 for x86_64-linux-gnu
system info: n_threads = 1, n_threads_batch = 1, total_threads = 8

system_info: n_threads = 1 (n_threads_batch = 1) / 8 | CPU : SSE3 = 1 | SSSE3 = 1 | AVX = 1 | AVX2 = 1 | F16C = 1 | FMA = 1 | BMI2 = 1 | LLAMAFILE = 1 | OPENMP = 1 | AARCH64_REPACK = 1 | 

Web UI is disabled
main: binding port with default address family
main: HTTP server is listening, hostname: 0.0.0.0, port: 80, http threads: 7
main: loading model
srv    load_model: loading model '/root/.cache/llama.cpp/nomic-ai_nomic-embed-text-v2-moe-GGUF_nomic-embed-text-v2-moe.Q4_K_M.gguf'
llama_model_loader: loaded meta data with 45 key-value pairs and 142 tensors from /root/.cache/llama.cpp/nomic-ai_nomic-embed-text-v2-moe-GGUF_nomic-embed-text-v2-moe.Q4_K_M.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              = nomic-bert-moe
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = nomic-embed-text-v2-moe
llama_model_loader: - kv   3:                            general.version str              = 2048
llama_model_loader: - kv   4:                       general.organization str              = Nomic Ai
llama_model_loader: - kv   5:                           general.basename str              = nomic-xlm
llama_model_loader: - kv   6:                         general.size_label str              = 8x277M
llama_model_loader: - kv   7:                            general.license str              = apache-2.0
llama_model_loader: - kv   8:                   general.base_model.count u32              = 1
llama_model_loader: - kv   9:                  general.base_model.0.name str              = Nomic Embed Text v2 Moe Unsupervised
llama_model_loader: - kv  10:          general.base_model.0.organization str              = Nomic Ai
llama_model_loader: - kv  11:              general.base_model.0.repo_url str              = https://huggingface.co/nomic-ai/nomic...
llama_model_loader: - kv  12:                               general.tags arr[str,4]       = ["sentence-transformers", "sentence-s...
llama_model_loader: - kv  13:                          general.languages arr[str,101]     = ["en", "es", "fr", "de", "it", "pt", ...
llama_model_loader: - kv  14:                 nomic-bert-moe.block_count u32              = 12
llama_model_loader: - kv  15:              nomic-bert-moe.context_length u32              = 512
llama_model_loader: - kv  16:            nomic-bert-moe.embedding_length u32              = 768
llama_model_loader: - kv  17:         nomic-bert-moe.feed_forward_length u32              = 3072
llama_model_loader: - kv  18:        nomic-bert-moe.attention.head_count u32              = 12
llama_model_loader: - kv  19: nomic-bert-moe.attention.layer_norm_epsilon f32              = 0.000010
llama_model_loader: - kv  20:            nomic-bert-moe.attention.causal bool             = false
llama_model_loader: - kv  21:                nomic-bert-moe.pooling_type u32              = 1
llama_model_loader: - kv  22:              nomic-bert-moe.rope.freq_base f32              = 10000.000000
llama_model_loader: - kv  23:          nomic-bert-moe.moe_every_n_layers u32              = 2
llama_model_loader: - kv  24:                nomic-bert-moe.expert_count u32              = 8
llama_model_loader: - kv  25:           nomic-bert-moe.expert_used_count u32              = 2
llama_model_loader: - kv  26:                       tokenizer.ggml.model str              = t5
llama_model_loader: - kv  27:                         tokenizer.ggml.pre str              = default
llama_model_loader: - kv  28:                      tokenizer.ggml.tokens arr[str,250048]  = ["<s>", "<pad>", "</s>", "<unk>", ","...
llama_model_loader: - kv  29:                      tokenizer.ggml.scores arr[f32,250048]  = [0.000000, 0.000000, 0.000000, 0.0000...
llama_model_loader: - kv  30:                  tokenizer.ggml.token_type arr[i32,250048]  = [3, 3, 3, 2, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  31:            tokenizer.ggml.add_space_prefix bool             = true
llama_model_loader: - kv  32:            tokenizer.ggml.token_type_count u32              = 1
llama_model_loader: - kv  33:    tokenizer.ggml.remove_extra_whitespaces bool             = true
llama_model_loader: - kv  34:        tokenizer.ggml.precompiled_charsmap arr[u8,237539]   = [0, 180, 2, 0, 0, 132, 0, 0, 0, 0, 0,...
llama_model_loader: - kv  35:                tokenizer.ggml.bos_token_id u32              = 0
llama_model_loader: - kv  36:                tokenizer.ggml.eos_token_id u32              = 2
llama_model_loader: - kv  37:            tokenizer.ggml.unknown_token_id u32              = 3
llama_model_loader: - kv  38:          tokenizer.ggml.seperator_token_id u32              = 2
llama_model_loader: - kv  39:            tokenizer.ggml.padding_token_id u32              = 1
llama_model_loader: - kv  40:               tokenizer.ggml.mask_token_id u32              = 250001
llama_model_loader: - kv  41:               tokenizer.ggml.add_bos_token bool             = true
llama_model_loader: - kv  42:               tokenizer.ggml.add_eos_token bool             = true
llama_model_loader: - kv  43:               general.quantization_version u32              = 2
llama_model_loader: - kv  44:                          general.file_type u32              = 15
llama_model_loader: - type  f32:   93 tensors
llama_model_loader: - type q4_K:   18 tensors
llama_model_loader: - type q5_K:   24 tensors
llama_model_loader: - type q6_K:    7 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 321.66 MiB (5.68 BPW) 
load: model vocab missing newline token, using special_pad_id instead
load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
load: special tokens cache size = 4
load: token to piece cache size = 2.1668 MB
print_info: arch             = nomic-bert-moe
print_info: vocab_only       = 0
print_info: n_ctx_train      = 512
print_info: n_embd           = 768
print_info: n_layer          = 12
print_info: n_head           = 12
print_info: n_head_kv        = 12
print_info: n_rot            = 64
print_info: n_swa            = 0
print_info: n_swa_pattern    = 1
print_info: n_embd_head_k    = 64
print_info: n_embd_head_v    = 64
print_info: n_gqa            = 1
print_info: n_embd_k_gqa     = 768
print_info: n_embd_v_gqa     = 768
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         = 8
print_info: n_expert_used    = 2
print_info: causal attn      = 0
print_info: pooling type     = 1
print_info: rope type        = 2
print_info: rope scaling     = linear
print_info: freq_base_train  = 10000.0
print_info: freq_scale_train = 1
print_info: n_ctx_orig_yarn  = 512
print_info: rope_finetuned   = unknown
print_info: ssm_d_conv       = 0
print_info: ssm_d_inner      = 0
print_info: ssm_d_state      = 0
print_info: ssm_dt_rank      = 0
print_info: ssm_dt_b_c_rms   = 0
print_info: model type       = 475M
print_info: model params     = 475.29 M
print_info: general.name     = nomic-embed-text-v2-moe
print_info: vocab type       = UGM
print_info: n_vocab          = 250048
print_info: n_merges         = 0
print_info: BOS token        = 0 '<s>'
print_info: EOS token        = 2 '</s>'
print_info: UNK token        = 3 '<unk>'
print_info: SEP token        = 2 '</s>'
print_info: PAD token        = 1 '<pad>'
print_info: MASK token       = 250001 '[PAD250000]'
print_info: LF token         = 0 '<s>'
print_info: EOG token        = 2 '</s>'
print_info: max token length = 48
load_tensors: loading model tensors, this can take a while... (mmap = true)
load_tensors:  CPU_AARCH64 model buffer size =   102.52 MiB
load_tensors:   CPU_Mapped model buffer size =   321.66 MiB
...........................
llama_context: constructing llama_context
llama_context: n_seq_max     = 4
llama_context: n_ctx         = 2048
llama_context: n_ctx_per_seq = 512
llama_context: n_batch       = 512
llama_context: n_ubatch      = 512
llama_context: causal_attn   = 0
llama_context: flash_attn    = 0
llama_context: freq_base     = 10000.0
llama_context: freq_scale    = 1
llama_context:        CPU  output buffer size =     0.00 MiB
common_init_from_params: KV cache shifting is not supported for this context, disabling KV cache shifting
common_init_from_params: setting dry_penalty_last_n to ctx_size = 2048
common_init_from_params: warming up the model with an empty run - please wait ... (--no-warmup to disable)
decode: cannot decode batches with this context (use llama_encode() instead)
srv          init: initializing slots, n_slots = 4
slot         init: id  0 | task -1 | new slot n_ctx_slot = 512
slot         init: id  1 | task -1 | new slot n_ctx_slot = 512
slot         init: id  2 | task -1 | new slot n_ctx_slot = 512
slot         init: id  3 | task -1 | new slot n_ctx_slot = 512
main: model loaded
main: chat template, chat_template: {%- for message in messages -%}
  {{- '<|im_start|>' + message.role + '
' + message.content + '<|im_end|>
' -}}
{%- endfor -%}
{%- if add_generation_prompt -%}
  {{- '<|im_start|>assistant
' -}}
{%- endif -%}, example_format: '<|im_start|>system
You are a helpful assistant<|im_end|>
<|im_start|>user
Hello<|im_end|>
<|im_start|>assistant
Hi there<|im_end|>
<|im_start|>user
How are you?<|im_end|>
<|im_start|>assistant
'
main: server is listening on http://0.0.0.0:80 - starting the main loop
slot launch_slot_: id  0 | task 499 | processing task
slot update_slots: id  0 | task 499 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 512
slot update_slots: id  0 | task 499 | kv cache rm [0, end)
slot update_slots: id  0 | task 499 | prompt processing progress, n_past = 512, n_tokens = 512, progress = 1.000000
slot update_slots: id  0 | task 499 | prompt done, n_past = 512, n_tokens = 512
slot      release: id  0 | task 499 | stop processing: n_past = 512, truncated = 0
srv  log_server_r: request: POST /v1/embeddings 10.2.0.77 200
slot launch_slot_: id  0 | task 526 | processing task
slot update_slots: id  0 | task 526 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 512
slot update_slots: id  0 | task 526 | kv cache rm [0, end)
slot update_slots: id  0 | task 526 | prompt processing progress, n_past = 512, n_tokens = 512, progress = 1.000000
slot update_slots: id  0 | task 526 | prompt done, n_past = 512, n_tokens = 512
slot      release: id  0 | task 526 | stop processing: n_past = 512, truncated = 0
srv  log_server_r: request: POST /v1/embeddings 10.2.0.77 200
slot launch_slot_: id  0 | task 1047 | processing task
slot update_slots: id  0 | task 1047 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 512
slot update_slots: id  0 | task 1047 | kv cache rm [0, end)
slot update_slots: id  0 | task 1047 | prompt processing progress, n_past = 512, n_tokens = 512, progress = 1.000000
slot update_slots: id  0 | task 1047 | prompt done, n_past = 512, n_tokens = 512
slot      release: id  0 | task 1047 | stop processing: n_past = 512, truncated = 0
srv  log_server_r: request: POST /v1/embeddings 10.2.0.104 200
slot launch_slot_: id  1 | task 1164 | processing task
slot update_slots: id  1 | task 1164 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 94
slot update_slots: id  1 | task 1164 | kv cache rm [0, end)
slot update_slots: id  1 | task 1164 | prompt processing progress, n_past = 94, n_tokens = 94, progress = 1.000000
slot update_slots: id  1 | task 1164 | prompt done, n_past = 94, n_tokens = 94
slot      release: id  1 | task 1164 | stop processing: n_past = 94, truncated = 0
srv  log_server_r: request: POST /v1/embeddings 10.2.0.104 200
slot launch_slot_: id  1 | task 1171 | processing task
slot update_slots: id  1 | task 1171 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 94
slot update_slots: id  1 | task 1171 | kv cache rm [0, end)
slot update_slots: id  1 | task 1171 | prompt processing progress, n_past = 94, n_tokens = 94, progress = 1.000000
slot update_slots: id  1 | task 1171 | prompt done, n_past = 94, n_tokens = 94
slot      release: id  1 | task 1171 | stop processing: n_past = 94, truncated = 0
srv  log_server_r: request: POST /v1/embeddings 10.2.0.104 200
slot launch_slot_: id  1 | task 1570 | processing task
slot update_slots: id  1 | task 1570 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 94
slot update_slots: id  1 | task 1570 | kv cache rm [0, end)
slot update_slots: id  1 | task 1570 | prompt processing progress, n_past = 94, n_tokens = 94, progress = 1.000000
slot update_slots: id  1 | task 1570 | prompt done, n_past = 94, n_tokens = 94
srv  log_server_r: request: POST /v1/embeddings 10.2.0.132 200
slot      release: id  1 | task 1570 | stop processing: n_past = 94, truncated = 0
slot launch_slot_: id  2 | task 2487 | processing task
slot update_slots: id  2 | task 2487 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 23
slot update_slots: id  2 | task 2487 | kv cache rm [0, end)
slot update_slots: id  2 | task 2487 | prompt processing progress, n_past = 23, n_tokens = 23, progress = 1.000000
slot update_slots: id  2 | task 2487 | prompt done, n_past = 23, n_tokens = 23
slot      release: id  2 | task 2487 | stop processing: n_past = 23, truncated = 0
srv  log_server_r: request: POST /v1/embeddings 10.2.0.132 200
slot launch_slot_: id  3 | task 2546 | processing task
slot launch_slot_: id  0 | task 2547 | processing task
slot launch_slot_: id  1 | task 2548 | processing task
slot launch_slot_: id  2 | task 2549 | processing task
slot update_slots: id  0 | task 2547 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 502
slot update_slots: id  0 | task 2547 | kv cache rm [0, end)
slot update_slots: id  0 | task 2547 | prompt processing progress, n_past = 502, n_tokens = 502, progress = 1.000000
slot update_slots: id  0 | task 2547 | prompt done, n_past = 502, n_tokens = 502
slot update_slots: id  1 | task 2548 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 502
slot update_slots: id  2 | task 2549 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 502
slot update_slots: id  3 | task 2546 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 501
slot      release: id  0 | task 2547 | stop processing: n_past = 502, truncated = 0
slot launch_slot_: id  0 | task 2550 | processing task
slot update_slots: id  0 | task 2550 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 2
slot update_slots: id  0 | task 2550 | kv cache rm [0, end)
slot update_slots: id  0 | task 2550 | prompt processing progress, n_past = 2, n_tokens = 2, progress = 1.000000
slot update_slots: id  0 | task 2550 | prompt done, n_past = 2, n_tokens = 2
slot update_slots: id  1 | task 2548 | kv cache rm [0, end)
slot update_slots: id  1 | task 2548 | prompt processing progress, n_past = 502, n_tokens = 504, progress = 1.000000
slot update_slots: id  1 | task 2548 | prompt done, n_past = 502, n_tokens = 504
slot      release: id  0 | task 2550 | stop processing: n_past = 2, truncated = 0
slot      release: id  1 | task 2548 | stop processing: n_past = 502, truncated = 0
slot update_slots: id  2 | task 2549 | kv cache rm [0, end)
slot update_slots: id  2 | task 2549 | prompt processing progress, n_past = 502, n_tokens = 502, progress = 1.000000
slot update_slots: id  2 | task 2549 | prompt done, n_past = 502, n_tokens = 502
slot      release: id  2 | task 2549 | stop processing: n_past = 502, truncated = 0
slot update_slots: id  3 | task 2546 | kv cache rm [0, end)
slot update_slots: id  3 | task 2546 | prompt processing progress, n_past = 501, n_tokens = 501, progress = 1.000000
slot update_slots: id  3 | task 2546 | prompt done, n_past = 501, n_tokens = 501
slot      release: id  3 | task 2546 | stop processing: n_past = 501, truncated = 0
srv  log_server_r: request: POST /v1/embeddings 10.2.0.132 200
slot launch_slot_: id  0 | task 2557 | processing task
slot update_slots: id  0 | task 2557 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 24
slot update_slots: id  0 | task 2557 | kv cache rm [0, end)
slot update_slots: id  0 | task 2557 | prompt processing progress, n_past = 24, n_tokens = 24, progress = 1.000000
slot update_slots: id  0 | task 2557 | prompt done, n_past = 24, n_tokens = 24
slot      release: id  0 | task 2557 | stop processing: n_past = 24, truncated = 0
srv  log_server_r: request: POST /v1/embeddings 10.2.0.132 200
slot launch_slot_: id  1 | task 2633 | processing task
slot update_slots: id  1 | task 2633 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 2
slot update_slots: id  1 | task 2633 | kv cache rm [0, end)
slot update_slots: id  1 | task 2633 | prompt processing progress, n_past = 2, n_tokens = 2, progress = 1.000000
slot update_slots: id  1 | task 2633 | prompt done, n_past = 2, n_tokens = 2
srv  log_server_r: request: POST /v1/embeddings 10.2.0.132 200
slot      release: id  1 | task 2633 | stop processing: n_past = 2, truncated = 0
slot launch_slot_: id  1 | task 2635 | processing task
slot update_slots: id  1 | task 2635 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 2
slot update_slots: id  1 | task 2635 | kv cache rm [0, end)
slot update_slots: id  1 | task 2635 | prompt processing progress, n_past = 2, n_tokens = 2, progress = 1.000000
slot update_slots: id  1 | task 2635 | prompt done, n_past = 2, n_tokens = 2
srv  log_server_r: request: POST /v1/embeddings 10.2.0.132 200
slot      release: id  1 | task 2635 | stop processing: n_past = 2, truncated = 0
slot launch_slot_: id  2 | task 2637 | processing task
slot update_slots: id  2 | task 2637 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 21
slot update_slots: id  2 | task 2637 | kv cache rm [0, end)
slot update_slots: id  2 | task 2637 | prompt processing progress, n_past = 21, n_tokens = 21, progress = 1.000000
slot update_slots: id  2 | task 2637 | prompt done, n_past = 21, n_tokens = 21
srv  log_server_r: request: POST /v1/embeddings 10.2.0.132 200
slot      release: id  2 | task 2637 | stop processing: n_past = 21, truncated = 0
slot launch_slot_: id  3 | task 11488 | processing task
slot update_slots: id  3 | task 11488 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 10
slot update_slots: id  3 | task 11488 | kv cache rm [0, end)
slot update_slots: id  3 | task 11488 | prompt processing progress, n_past = 10, n_tokens = 10, progress = 1.000000
slot update_slots: id  3 | task 11488 | prompt done, n_past = 10, n_tokens = 10
slot      release: id  3 | task 11488 | stop processing: n_past = 10, truncated = 0
srv  log_server_r: request: POST /v1/embeddings 10.2.0.132 200
slot launch_slot_: id  3 | task 11513 | processing task
slot update_slots: id  3 | task 11513 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 10
slot update_slots: id  3 | task 11513 | kv cache rm [0, end)
slot update_slots: id  3 | task 11513 | prompt processing progress, n_past = 10, n_tokens = 10, progress = 1.000000
slot update_slots: id  3 | task 11513 | prompt done, n_past = 10, n_tokens = 10
slot      release: id  3 | task 11513 | stop processing: n_past = 10, truncated = 0
srv  log_server_r: request: POST /v1/embeddings 10.2.0.132 200
slot launch_slot_: id  0 | task 11524 | processing task
slot update_slots: id  0 | task 11524 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 12
slot update_slots: id  0 | task 11524 | kv cache rm [0, end)
slot update_slots: id  0 | task 11524 | prompt processing progress, n_past = 12, n_tokens = 12, progress = 1.000000
slot update_slots: id  0 | task 11524 | prompt done, n_past = 12, n_tokens = 12
slot      release: id  0 | task 11524 | stop processing: n_past = 12, truncated = 0
srv  log_server_r: request: POST /v1/embeddings 10.2.0.132 200
slot launch_slot_: id  1 | task 11547 | processing task
slot launch_slot_: id  2 | task 11548 | processing task
slot update_slots: id  1 | task 11547 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 502
slot update_slots: id  1 | task 11547 | kv cache rm [0, end)
slot update_slots: id  1 | task 11547 | prompt processing progress, n_past = 502, n_tokens = 502, progress = 1.000000
slot update_slots: id  1 | task 11547 | prompt done, n_past = 502, n_tokens = 502
slot update_slots: id  2 | task 11548 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 2
slot update_slots: id  2 | task 11548 | kv cache rm [0, end)
slot update_slots: id  2 | task 11548 | prompt processing progress, n_past = 2, n_tokens = 504, progress = 1.000000
slot update_slots: id  2 | task 11548 | prompt done, n_past = 2, n_tokens = 504
slot      release: id  1 | task 11547 | stop processing: n_past = 502, truncated = 0
slot      release: id  2 | task 11548 | stop processing: n_past = 2, truncated = 0
slot launch_slot_: id  3 | task 11550 | processing task
slot launch_slot_: id  0 | task 11551 | processing task
slot launch_slot_: id  1 | task 11552 | processing task
slot launch_slot_: id  2 | task 11553 | processing task
slot update_slots: id  0 | task 11551 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 502
slot update_slots: id  0 | task 11551 | kv cache rm [0, end)
slot update_slots: id  0 | task 11551 | prompt processing progress, n_past = 502, n_tokens = 502, progress = 1.000000
slot update_slots: id  0 | task 11551 | prompt done, n_past = 502, n_tokens = 502
slot update_slots: id  1 | task 11552 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 502
slot update_slots: id  2 | task 11553 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 503
slot update_slots: id  3 | task 11550 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 501
srv  log_server_r: request: POST /v1/embeddings 10.2.0.132 200
slot      release: id  0 | task 11551 | stop processing: n_past = 502, truncated = 0
slot launch_slot_: id  0 | task 11554 | processing task
slot update_slots: id  0 | task 11554 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 502
slot update_slots: id  0 | task 11554 | kv cache rm [0, end)
slot update_slots: id  0 | task 11554 | prompt processing progress, n_past = 502, n_tokens = 502, progress = 1.000000
slot update_slots: id  0 | task 11554 | prompt done, n_past = 502, n_tokens = 502
slot      release: id  0 | task 11554 | stop processing: n_past = 502, truncated = 0
slot update_slots: id  1 | task 11552 | kv cache rm [0, end)
slot update_slots: id  1 | task 11552 | prompt processing progress, n_past = 502, n_tokens = 502, progress = 1.000000
slot update_slots: id  1 | task 11552 | prompt done, n_past = 502, n_tokens = 502
slot      release: id  1 | task 11552 | stop processing: n_past = 502, truncated = 0
slot update_slots: id  2 | task 11553 | kv cache rm [0, end)
slot update_slots: id  2 | task 11553 | prompt processing progress, n_past = 503, n_tokens = 503, progress = 1.000000
slot update_slots: id  2 | task 11553 | prompt done, n_past = 503, n_tokens = 503
slot      release: id  2 | task 11553 | stop processing: n_past = 503, truncated = 0
slot update_slots: id  3 | task 11550 | kv cache rm [0, end)
slot update_slots: id  3 | task 11550 | prompt processing progress, n_past = 501, n_tokens = 501, progress = 1.000000
slot update_slots: id  3 | task 11550 | prompt done, n_past = 501, n_tokens = 501
slot      release: id  3 | task 11550 | stop processing: n_past = 501, truncated = 0
srv  log_server_r: request: POST /v1/embeddings 10.2.0.132 200
slot launch_slot_: id  0 | task 11595 | processing task
slot update_slots: id  0 | task 11595 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 2
slot update_slots: id  0 | task 11595 | kv cache rm [0, end)
slot update_slots: id  0 | task 11595 | prompt processing progress, n_past = 2, n_tokens = 2, progress = 1.000000
slot update_slots: id  0 | task 11595 | prompt done, n_past = 2, n_tokens = 2
slot      release: id  0 | task 11595 | stop processing: n_past = 2, truncated = 0
srv  log_server_r: request: POST /v1/embeddings 10.2.0.132 200
slot launch_slot_: id  3 | task 11597 | processing task
slot launch_slot_: id  1 | task 11598 | processing task
slot launch_slot_: id  2 | task 11599 | processing task
slot launch_slot_: id  0 | task 11600 | processing task
slot update_slots: id  0 | task 11600 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 503
slot update_slots: id  0 | task 11600 | kv cache rm [0, end)
slot update_slots: id  0 | task 11600 | prompt processing progress, n_past = 503, n_tokens = 503, progress = 1.000000
slot update_slots: id  0 | task 11600 | prompt done, n_past = 503, n_tokens = 503
slot update_slots: id  1 | task 11598 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 502
slot update_slots: id  2 | task 11599 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 502
slot update_slots: id  3 | task 11597 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 501
slot      release: id  0 | task 11600 | stop processing: n_past = 503, truncated = 0
slot launch_slot_: id  0 | task 11603 | processing task
slot update_slots: id  0 | task 11603 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 502
slot update_slots: id  0 | task 11603 | kv cache rm [0, end)
slot update_slots: id  0 | task 11603 | prompt processing progress, n_past = 502, n_tokens = 502, progress = 1.000000
slot update_slots: id  0 | task 11603 | prompt done, n_past = 502, n_tokens = 502
slot      release: id  0 | task 11603 | stop processing: n_past = 502, truncated = 0
slot launch_slot_: id  0 | task 11604 | processing task
slot update_slots: id  0 | task 11604 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 2
slot update_slots: id  0 | task 11604 | kv cache rm [0, end)
slot update_slots: id  0 | task 11604 | prompt processing progress, n_past = 2, n_tokens = 2, progress = 1.000000
slot update_slots: id  0 | task 11604 | prompt done, n_past = 2, n_tokens = 2
slot update_slots: id  1 | task 11598 | kv cache rm [0, end)
slot update_slots: id  1 | task 11598 | prompt processing progress, n_past = 502, n_tokens = 504, progress = 1.000000
slot update_slots: id  1 | task 11598 | prompt done, n_past = 502, n_tokens = 504
slot      release: id  0 | task 11604 | stop processing: n_past = 2, truncated = 0
slot      release: id  1 | task 11598 | stop processing: n_past = 502, truncated = 0
slot launch_slot_: id  0 | task 11601 | processing task
slot update_slots: id  0 | task 11601 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 502
slot update_slots: id  0 | task 11601 | kv cache rm [0, end)
slot update_slots: id  0 | task 11601 | prompt processing progress, n_past = 502, n_tokens = 502, progress = 1.000000
slot update_slots: id  0 | task 11601 | prompt done, n_past = 502, n_tokens = 502
srv  log_server_r: request: POST /v1/embeddings 10.2.0.132 200
slot      release: id  0 | task 11601 | stop processing: n_past = 502, truncated = 0
slot update_slots: id  2 | task 11599 | kv cache rm [0, end)
slot update_slots: id  2 | task 11599 | prompt processing progress, n_past = 502, n_tokens = 502, progress = 1.000000
slot update_slots: id  2 | task 11599 | prompt done, n_past = 502, n_tokens = 502
slot      release: id  2 | task 11599 | stop processing: n_past = 502, truncated = 0
slot update_slots: id  3 | task 11597 | kv cache rm [0, end)
slot update_slots: id  3 | task 11597 | prompt processing progress, n_past = 501, n_tokens = 501, progress = 1.000000
slot update_slots: id  3 | task 11597 | prompt done, n_past = 501, n_tokens = 501
slot      release: id  3 | task 11597 | stop processing: n_past = 501, truncated = 0
srv  log_server_r: request: POST /v1/embeddings 10.2.0.132 200
slot launch_slot_: id  1 | task 11613 | processing task
slot launch_slot_: id  0 | task 11614 | processing task
slot launch_slot_: id  2 | task 11615 | processing task
slot update_slots: id  0 | task 11614 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 503
slot update_slots: id  0 | task 11614 | kv cache rm [0, end)
slot update_slots: id  0 | task 11614 | prompt processing progress, n_past = 503, n_tokens = 503, progress = 1.000000
slot update_slots: id  0 | task 11614 | prompt done, n_past = 503, n_tokens = 503
slot update_slots: id  1 | task 11613 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 501
slot update_slots: id  2 | task 11615 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 2
slot update_slots: id  2 | task 11615 | kv cache rm [0, end)
slot update_slots: id  2 | task 11615 | prompt processing progress, n_past = 2, n_tokens = 505, progress = 1.000000
slot update_slots: id  2 | task 11615 | prompt done, n_past = 2, n_tokens = 505
slot      release: id  0 | task 11614 | stop processing: n_past = 503, truncated = 0
slot      release: id  2 | task 11615 | stop processing: n_past = 2, truncated = 0
slot update_slots: id  1 | task 11613 | kv cache rm [0, end)
slot update_slots: id  1 | task 11613 | prompt processing progress, n_past = 501, n_tokens = 501, progress = 1.000000
slot update_slots: id  1 | task 11613 | prompt done, n_past = 501, n_tokens = 501
slot      release: id  1 | task 11613 | stop processing: n_past = 501, truncated = 0
srv  log_server_r: request: POST /v1/embeddings 10.2.0.132 200
slot launch_slot_: id  3 | task 11619 | processing task
slot update_slots: id  3 | task 11619 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 12
slot update_slots: id  3 | task 11619 | kv cache rm [0, end)
slot update_slots: id  3 | task 11619 | prompt processing progress, n_past = 12, n_tokens = 12, progress = 1.000000
slot update_slots: id  3 | task 11619 | prompt done, n_past = 12, n_tokens = 12
slot      release: id  3 | task 11619 | stop processing: n_past = 12, truncated = 0
srv  log_server_r: request: POST /v1/embeddings 10.2.0.132 200
slot launch_slot_: id  2 | task 11647 | processing task
slot update_slots: id  2 | task 11647 | new prompt, n_ctx_slot = 512, n_keep = 0, n_prompt_tokens = 2
slot update_slots: id  2 | task 11647 | kv cache rm [0, end)
slot update_slots: id  2 | task 11647 | prompt processing progress, n_past = 2, n_tokens = 2, progress = 1.000000
slot update_slots: id  2 | task 11647 | prompt done, n_past = 2, n_tokens = 2
/app/src/llama-graph.cpp:185: GGML_ASSERT(seq_id < n_tokens && "seq_id cannot be larger than n_tokens with pooling_type == MEAN") failed
srv  cancel_tasks: cancel task, id_task = 11647
srv  log_server_r: request: POST /v1/embeddings 10.2.0.132 200

Note: It is not deterministic, but it seems to happen more frequently when enough slots are used. If wanting to reproduce, you should reduce LLAMA_ARG_N_PARALLEL to 2, for instance.