build: 3785 (64c6af31) with cc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0 for x86_64-linux-gnu llama_model_loader: loaded meta data with 34 key-value pairs and 579 tensors from Qwen2.5-14B-Instruct-IMat-GGUF/Qwen2.5-14B-Instruct.Q8_0.gguf.hardlink.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 = qwen2 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = Qwen2.5 14B Instruct llama_model_loader: - kv 3: general.finetune str = Instruct llama_model_loader: - kv 4: general.basename str = Qwen2.5 llama_model_loader: - kv 5: general.size_label str = 14B llama_model_loader: - kv 6: general.license str = apache-2.0 llama_model_loader: - kv 7: general.license.link str = https://huggingface.co/Qwen/Qwen2.5-1... llama_model_loader: - kv 8: general.base_model.count u32 = 1 llama_model_loader: - kv 9: general.base_model.0.name str = Qwen2.5 14B llama_model_loader: - kv 10: general.base_model.0.organization str = Qwen llama_model_loader: - kv 11: general.base_model.0.repo_url str = https://huggingface.co/Qwen/Qwen2.5-14B llama_model_loader: - kv 12: general.tags arr[str,2] = ["chat", "text-generation"] llama_model_loader: - kv 13: general.languages arr[str,1] = ["en"] llama_model_loader: - kv 14: qwen2.block_count u32 = 48 llama_model_loader: - kv 15: qwen2.context_length u32 = 32768 llama_model_loader: - kv 16: qwen2.embedding_length u32 = 5120 llama_model_loader: - kv 17: qwen2.feed_forward_length u32 = 13824 llama_model_loader: - kv 18: qwen2.attention.head_count u32 = 40 llama_model_loader: - kv 19: qwen2.attention.head_count_kv u32 = 8 llama_model_loader: - kv 20: qwen2.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 21: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 22: general.file_type u32 = 7 llama_model_loader: - kv 23: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 24: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 25: tokenizer.ggml.tokens arr[str,152064] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 26: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 27: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 28: tokenizer.ggml.eos_token_id u32 = 151645 llama_model_loader: - kv 29: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 30: tokenizer.ggml.bos_token_id u32 = 151643 llama_model_loader: - kv 31: tokenizer.ggml.add_bos_token bool = false llama_model_loader: - kv 32: tokenizer.chat_template str = {%- if tools %}\n {{- '<|im_start|>... llama_model_loader: - kv 33: general.quantization_version u32 = 2 llama_model_loader: - type f32: 241 tensors llama_model_loader: - type q8_0: 338 tensors llm_load_vocab: special tokens cache size = 22 llm_load_vocab: token to piece cache size = 0.9310 MB llm_load_print_meta: format = GGUF V3 (latest) llm_load_print_meta: arch = qwen2 llm_load_print_meta: vocab type = BPE llm_load_print_meta: n_vocab = 152064 llm_load_print_meta: n_merges = 151387 llm_load_print_meta: vocab_only = 0 llm_load_print_meta: n_ctx_train = 32768 llm_load_print_meta: n_embd = 5120 llm_load_print_meta: n_layer = 48 llm_load_print_meta: n_head = 40 llm_load_print_meta: n_head_kv = 8 llm_load_print_meta: n_rot = 128 llm_load_print_meta: n_swa = 0 llm_load_print_meta: n_embd_head_k = 128 llm_load_print_meta: n_embd_head_v = 128 llm_load_print_meta: n_gqa = 5 llm_load_print_meta: n_embd_k_gqa = 1024 llm_load_print_meta: n_embd_v_gqa = 1024 llm_load_print_meta: f_norm_eps = 0.0e+00 llm_load_print_meta: f_norm_rms_eps = 1.0e-06 llm_load_print_meta: f_clamp_kqv = 0.0e+00 llm_load_print_meta: f_max_alibi_bias = 0.0e+00 llm_load_print_meta: f_logit_scale = 0.0e+00 llm_load_print_meta: n_ff = 13824 llm_load_print_meta: n_expert = 0 llm_load_print_meta: n_expert_used = 0 llm_load_print_meta: causal attn = 1 llm_load_print_meta: pooling type = 0 llm_load_print_meta: rope type = 2 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_ctx_orig_yarn = 32768 llm_load_print_meta: rope_finetuned = unknown llm_load_print_meta: ssm_d_conv = 0 llm_load_print_meta: ssm_d_inner = 0 llm_load_print_meta: ssm_d_state = 0 llm_load_print_meta: ssm_dt_rank = 0 llm_load_print_meta: ssm_dt_b_c_rms = 0 llm_load_print_meta: model type = ?B llm_load_print_meta: model ftype = Q8_0 llm_load_print_meta: model params = 14.77 B llm_load_print_meta: model size = 14.62 GiB (8.50 BPW) llm_load_print_meta: general.name = Qwen2.5 14B Instruct llm_load_print_meta: BOS token = 151643 '<|endoftext|>' llm_load_print_meta: EOS token = 151645 '<|im_end|>' llm_load_print_meta: PAD token = 151643 '<|endoftext|>' llm_load_print_meta: LF token = 148848 'ÄĬ' llm_load_print_meta: EOT token = 151645 '<|im_end|>' llm_load_print_meta: max token length = 256 ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no ggml_cuda_init: found 1 CUDA devices: Device 0: NVIDIA GeForce RTX 4090, compute capability 8.9, VMM: yes llm_load_tensors: ggml ctx size = 0.51 MiB llm_load_tensors: offloading 43 repeating layers to GPU llm_load_tensors: offloaded 43/49 layers to GPU llm_load_tensors: CPU buffer size = 14968.52 MiB llm_load_tensors: CUDA0 buffer size = 11995.82 MiB ............................................................................................ llama_new_context_with_model: n_ctx = 512 llama_new_context_with_model: n_batch = 512 llama_new_context_with_model: n_ubatch = 512 llama_new_context_with_model: flash_attn = 0 llama_new_context_with_model: freq_base = 1000000.0 llama_new_context_with_model: freq_scale = 1 llama_kv_cache_init: CUDA_Host KV buffer size = 10.00 MiB llama_kv_cache_init: CUDA0 KV buffer size = 86.00 MiB llama_new_context_with_model: KV self size = 96.00 MiB, K (f16): 48.00 MiB, V (f16): 48.00 MiB llama_new_context_with_model: CUDA_Host output buffer size = 0.58 MiB llama_new_context_with_model: CUDA0 compute buffer size = 1095.91 MiB llama_new_context_with_model: CUDA_Host compute buffer size = 11.01 MiB llama_new_context_with_model: graph nodes = 1686 llama_new_context_with_model: graph splits = 74 system_info: n_threads = 25 (n_threads_batch = 25) / 32 | AVX = 1 | AVX_VNNI = 0 | AVX2 = 1 | AVX512 = 1 | AVX512_VBMI = 1 | AVX512_VNNI = 1 | AVX512_BF16 = 1 | FMA = 1 | NEON = 0 | SVE = 0 | ARM_FMA = 0 | F16C = 1 | FP16_VA = 0 | RISCV_VECT = 0 | WASM_SIMD = 0 | BLAS = 1 | SSE3 = 1 | SSSE3 = 1 | VSX = 0 | MATMUL_INT8 = 0 | LLAMAFILE = 1 | compute_imatrix: tokenizing the input .. compute_imatrix: tokenization took 134.934 ms compute_imatrix: computing over 128 chunks with batch_size 512 compute_imatrix: 1.49 seconds per pass - ETA 3.18 minutes [1]4.3220,[2]3.0683,[3]2.9854,[4]3.3406,[5]3.2835,[6]3.0286,[7]3.2552,[8]3.2639,[9]3.6543,[10]3.6131,[11]3.5877,[12]3.9079,[13]4.3486,[14]4.5803,[15]4.9686,[16]5.2464,[17]5.4390,[18]5.7663,[19]5.5970,[20]5.7251,[21]5.7153,[22]5.7635,[23]5.6519,[24]5.8260,[25]5.9839,[26]5.8727,[27]5.7475,[28]5.5062,[29]5.4094,[30]5.3970,[31]5.2606,[32]5.0918,[33]5.0093,[34]4.9394,[35]4.9087,[36]4.9043,[37]4.9178,[38]4.9579,[39]4.9499,[40]5.0846,[41]5.1395,[42]5.1632,[43]5.0847,[44]5.1808,[45]5.2019,[46]5.2780,[47]5.2234,[48]5.2869,[49]5.3759,[50]5.4509,[51]5.3891,[52]5.4642,[53]5.6054,[54]5.6880,[55]5.7635,[56]5.8496,[57]5.9086,[58]5.9634,[59]6.0047,[60]6.0386,[61]6.0314,[62]6.0119,[63]6.0579,[64]6.1157,[65]6.0773,[66]6.0841,[67]6.0965,[68]6.0437,[69]6.0157,[70]6.0115,[71]5.9981,[72]5.9987,[73]6.0158,[74]5.9719,[75]5.9345,[76]5.8992,[77]5.8880,[78]5.8776,[79]5.8613,[80]5.8116,[81]5.8382,[82]5.8386,[83]5.8102,[84]5.8262,[85]5.8370,[86]5.8071,[87]5.7921,[88]5.7834,[89]5.8069,[90]5.8339,[91]5.8343,[92]5.8076,[93]5.7806,[94]5.7416,[95]5.7114,[96]5.6846,[97]5.6536,[98]5.6239,[99]5.6096,[100]5.6227,[101]5.6509,[102]5.7260,[103]5.8009,[104]5.8591,[105]5.9567,[106]6.0257,[107]6.0575,[108]6.0531,[109]6.0503,[110]6.0368,[111]5.9941,[112]5.9334,[113]5.8905,[114]5.9341,[115]5.9453,[116]5.9565,[117]5.9773,[118]6.0121,[119]6.0182,[120]6.0192,[121]6.0384,[122]6.0133,[123]6.0451,[124]6.0549,[125]6.0672,[126]6.0816,[127]6.0860,[128]6.1058, Final estimate: PPL = 6.1058 +/- 0.08196 llama_perf_context_print: load time = 4036.12 ms llama_perf_context_print: prompt eval time = 142821.59 ms / 65536 tokens ( 2.18 ms per token, 458.87 tokens per second) llama_perf_context_print: eval time = 0.00 ms / 1 runs ( 0.00 ms per token, inf tokens per second) llama_perf_context_print: total time = 146405.08 ms / 65537 tokens