llama_model_loader: loaded meta data with 25 key-value pairs and 435 tensors from Yi-9B-Coder-IMat-GGUF/Yi-9B-Coder.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 = llama llama_model_loader: - kv 1: general.name str = Yi-9B-Coder llama_model_loader: - kv 2: llama.block_count u32 = 48 llama_model_loader: - kv 3: llama.context_length u32 = 4096 llama_model_loader: - kv 4: llama.embedding_length u32 = 4096 llama_model_loader: - kv 5: llama.feed_forward_length u32 = 11008 llama_model_loader: - kv 6: llama.attention.head_count u32 = 32 llama_model_loader: - kv 7: llama.attention.head_count_kv u32 = 4 llama_model_loader: - kv 8: llama.rope.freq_base f32 = 10000.000000 llama_model_loader: - kv 9: llama.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 10: general.file_type u32 = 7 llama_model_loader: - kv 11: llama.vocab_size u32 = 64000 llama_model_loader: - kv 12: llama.rope.dimension_count u32 = 128 llama_model_loader: - kv 13: tokenizer.ggml.model str = llama llama_model_loader: - kv 14: tokenizer.ggml.pre str = default llama_model_loader: - kv 15: tokenizer.ggml.tokens arr[str,64000] = ["", "<|startoftext|>", "<|endof... llama_model_loader: - kv 16: tokenizer.ggml.scores arr[f32,64000] = [0.000000, 0.000000, 0.000000, 0.0000... llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,64000] = [2, 3, 3, 3, 3, 3, 1, 1, 1, 3, 3, 3, ... llama_model_loader: - kv 18: tokenizer.ggml.bos_token_id u32 = 1 llama_model_loader: - kv 19: tokenizer.ggml.eos_token_id u32 = 2 llama_model_loader: - kv 20: tokenizer.ggml.unknown_token_id u32 = 0 llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 2 llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = false llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 24: general.quantization_version u32 = 2 llama_model_loader: - type f32: 97 tensors llama_model_loader: - type q8_0: 338 tensors llm_load_vocab: special tokens cache size = 267 llm_load_vocab: token to piece cache size = 0.3834 MB llm_load_print_meta: format = GGUF V3 (latest) llm_load_print_meta: arch = llama llm_load_print_meta: vocab type = SPM llm_load_print_meta: n_vocab = 64000 llm_load_print_meta: n_merges = 0 llm_load_print_meta: n_ctx_train = 4096 llm_load_print_meta: n_embd = 4096 llm_load_print_meta: n_head = 32 llm_load_print_meta: n_head_kv = 4 llm_load_print_meta: n_layer = 48 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 = 8 llm_load_print_meta: n_embd_k_gqa = 512 llm_load_print_meta: n_embd_v_gqa = 512 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 = 11008 llm_load_print_meta: n_expert = 0 llm_load_print_meta: n_expert_used = 0 llm_load_print_meta: causal attn = 1 llm_load_print_meta: pooling type = 0 llm_load_print_meta: rope type = 0 llm_load_print_meta: rope scaling = linear llm_load_print_meta: freq_base_train = 10000.0 llm_load_print_meta: freq_scale_train = 1 llm_load_print_meta: n_ctx_orig_yarn = 4096 llm_load_print_meta: rope_finetuned = unknown llm_load_print_meta: ssm_d_conv = 0 llm_load_print_meta: ssm_d_inner = 0 llm_load_print_meta: ssm_d_state = 0 llm_load_print_meta: ssm_dt_rank = 0 llm_load_print_meta: model type = 34B llm_load_print_meta: model ftype = Q8_0 llm_load_print_meta: model params = 8.83 B llm_load_print_meta: model size = 8.74 GiB (8.50 BPW) llm_load_print_meta: general.name = Yi-9B-Coder llm_load_print_meta: BOS token = 1 '<|startoftext|>' llm_load_print_meta: EOS token = 2 '<|endoftext|>' llm_load_print_meta: UNK token = 0 '' llm_load_print_meta: PAD token = 2 '<|endoftext|>' llm_load_print_meta: LF token = 315 '<0x0A>' llm_load_print_meta: EOT token = 2 '<|endoftext|>' llm_load_print_meta: max token length = 48 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 = 0.41 MiB llm_load_tensors: offloading 48 repeating layers to GPU llm_load_tensors: offloading non-repeating layers to GPU llm_load_tensors: offloaded 49/49 layers to GPU llm_load_tensors: CPU buffer size = 265.62 MiB llm_load_tensors: CUDA0 buffer size = 8682.16 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 = 10000.0 llama_new_context_with_model: freq_scale = 1 llama_kv_cache_init: CUDA0 KV buffer size = 48.00 MiB llama_new_context_with_model: KV self size = 48.00 MiB, K (f16): 24.00 MiB, V (f16): 24.00 MiB llama_new_context_with_model: CUDA_Host output buffer size = 0.24 MiB llama_new_context_with_model: CUDA0 compute buffer size = 133.00 MiB llama_new_context_with_model: CUDA_Host compute buffer size = 9.01 MiB llama_new_context_with_model: graph nodes = 1542 llama_new_context_with_model: graph splits = 2 system_info: n_threads = 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 | 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 99.517 ms compute_imatrix: computing over 146 chunks with batch_size 512 compute_imatrix: 0.84 seconds per pass - ETA 2.05 minutes [1]23.1631,[2]14.0201,[3]13.8337,[4]15.6910,[5]15.9979,[6]16.4981,[7]12.7437,[8]15.7290,[9]15.6026, save_imatrix: stored collected data after 10 chunks in Yi-9B-Coder-IMat-GGUF/imatrix.dat [10]17.7123,[11]17.9431,[12]16.1416,[13]17.4802,[14]19.1233,[15]20.4605,[16]22.1210,[17]23.6353,[18]23.7434,[19]24.2365, save_imatrix: stored collected data after 20 chunks in Yi-9B-Coder-IMat-GGUF/imatrix.dat [20]25.1431,[21]23.3290,[22]23.0257,[23]23.8386,[24]24.4275,[25]24.5770,[26]23.8805,[27]24.7346,[28]25.5361,[29]26.9611, save_imatrix: stored collected data after 30 chunks in Yi-9B-Coder-IMat-GGUF/imatrix.dat [30]27.2252,[31]27.8791,[32]28.8940,[33]28.4610,[34]29.5878,[35]29.8153,[36]27.7273,[37]25.7733,[38]25.4580,[39]25.1857, save_imatrix: stored collected data after 40 chunks in Yi-9B-Coder-IMat-GGUF/imatrix.dat [40]24.8920,[41]24.4668,[42]23.8280,[43]23.6563,[44]23.0919,[45]22.8261,[46]22.7168,[47]22.9998,[48]23.3655,[49]23.7772, save_imatrix: stored collected data after 50 chunks in Yi-9B-Coder-IMat-GGUF/imatrix.dat [50]24.3145,[51]25.3137,[52]26.3264,[53]26.8731,[54]27.2882,[55]27.3431,[56]26.8452,[57]27.3450,[58]27.7474,[59]27.9772, save_imatrix: stored collected data after 60 chunks in Yi-9B-Coder-IMat-GGUF/imatrix.dat [60]27.5615,[61]27.1916,[62]27.2339,[63]27.6864,[64]28.2476,[65]28.7021,[66]28.8739,[67]29.1852,[68]29.3976,[69]29.5442, save_imatrix: stored collected data after 70 chunks in Yi-9B-Coder-IMat-GGUF/imatrix.dat [70]29.0318,[71]28.6331,[72]28.2458,[73]27.9526,[74]28.1002,[75]28.2634,[76]28.2284,[77]28.1927,[78]28.0400,[79]27.7259, save_imatrix: stored collected data after 80 chunks in Yi-9B-Coder-IMat-GGUF/imatrix.dat [80]27.4541,[81]27.0402,[82]27.0662,[83]26.8777,[84]26.6661,[85]26.6754,[86]26.4203,[87]26.1359,[88]25.9176,[89]25.9078, save_imatrix: stored collected data after 90 chunks in Yi-9B-Coder-IMat-GGUF/imatrix.dat [90]25.7199,[91]25.6668,[92]25.2476,[93]25.1993,[94]25.3737,[95]25.4112,[96]25.1537,[97]25.2050,[98]25.2597,[99]25.2710, save_imatrix: stored collected data after 100 chunks in Yi-9B-Coder-IMat-GGUF/imatrix.dat [100]24.8283,[101]24.8705,[102]24.7595,[103]24.8406,[104]24.8942,[105]25.0121,[106]24.7367,[107]24.4717,[108]24.1997,[109]23.9286, save_imatrix: stored collected data after 110 chunks in Yi-9B-Coder-IMat-GGUF/imatrix.dat [110]23.6729,[111]23.4221,[112]23.2042,[113]22.9676,[114]22.9006,[115]22.9633,[116]23.1386,[117]23.4191,[118]23.6960,[119]23.9039, save_imatrix: stored collected data after 120 chunks in Yi-9B-Coder-IMat-GGUF/imatrix.dat [120]24.3006,[121]24.6783,[122]24.8009,[123]24.8656,[124]24.6197,[125]24.6086,[126]24.4142,[127]24.3009,[128]24.1589,[129]24.2113, save_imatrix: stored collected data after 130 chunks in Yi-9B-Coder-IMat-GGUF/imatrix.dat [130]24.3836,[131]24.4772,[132]24.5623,[133]24.6763,[134]24.8778,[135]24.9183,[136]24.9277,[137]24.9997,[138]25.0110,[139]24.9703, save_imatrix: stored collected data after 140 chunks in Yi-9B-Coder-IMat-GGUF/imatrix.dat [140]25.1931,[141]25.3727,[142]25.5644,[143]25.8610,[144]26.1739,[145]26.4005,[146]26.6039, save_imatrix: stored collected data after 146 chunks in Yi-9B-Coder-IMat-GGUF/imatrix.dat llama_print_timings: load time = 2143.24 ms llama_print_timings: sample time = 0.00 ms / 1 runs ( 0.00 ms per token, inf tokens per second) llama_print_timings: prompt eval time = 113628.93 ms / 74752 tokens ( 1.52 ms per token, 657.86 tokens per second) llama_print_timings: eval time = 0.00 ms / 1 runs ( 0.00 ms per token, inf tokens per second) llama_print_timings: total time = 115527.76 ms / 74753 tokens Final estimate: PPL = 26.6039 +/- 0.59861