llama_model_loader: loaded meta data with 39 key-value pairs and 959 tensors from DeepSeek-Coder-V2-Instruct-IMat-GGUF/DeepSeek-Coder-V2-Instruct.Q8_0.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 = deepseek2 llama_model_loader: - kv 1: general.name str = DeepSeek-Coder-V2-Instruct llama_model_loader: - kv 2: deepseek2.block_count u32 = 60 llama_model_loader: - kv 3: deepseek2.context_length u32 = 163840 llama_model_loader: - kv 4: deepseek2.embedding_length u32 = 5120 llama_model_loader: - kv 5: deepseek2.feed_forward_length u32 = 12288 llama_model_loader: - kv 6: deepseek2.attention.head_count u32 = 128 llama_model_loader: - kv 7: deepseek2.attention.head_count_kv u32 = 128 llama_model_loader: - kv 8: deepseek2.rope.freq_base f32 = 10000.000000 llama_model_loader: - kv 9: deepseek2.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 10: deepseek2.expert_used_count u32 = 6 llama_model_loader: - kv 11: general.file_type u32 = 7 llama_model_loader: - kv 12: deepseek2.leading_dense_block_count u32 = 1 llama_model_loader: - kv 13: deepseek2.vocab_size u32 = 102400 llama_model_loader: - kv 14: deepseek2.attention.q_lora_rank u32 = 1536 llama_model_loader: - kv 15: deepseek2.attention.kv_lora_rank u32 = 512 llama_model_loader: - kv 16: deepseek2.attention.key_length u32 = 192 llama_model_loader: - kv 17: deepseek2.attention.value_length u32 = 128 llama_model_loader: - kv 18: deepseek2.expert_feed_forward_length u32 = 1536 llama_model_loader: - kv 19: deepseek2.expert_count u32 = 160 llama_model_loader: - kv 20: deepseek2.expert_shared_count u32 = 2 llama_model_loader: - kv 21: deepseek2.expert_weights_scale f32 = 16.000000 llama_model_loader: - kv 22: deepseek2.rope.dimension_count u32 = 64 llama_model_loader: - kv 23: deepseek2.rope.scaling.type str = yarn llama_model_loader: - kv 24: deepseek2.rope.scaling.factor f32 = 40.000000 llama_model_loader: - kv 25: deepseek2.rope.scaling.original_context_length u32 = 4096 llama_model_loader: - kv 26: deepseek2.rope.scaling.yarn_log_multiplier f32 = 0.100000 llama_model_loader: - kv 27: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 28: tokenizer.ggml.pre str = deepseek-llm llama_model_loader: - kv 29: tokenizer.ggml.tokens arr[str,102400] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 30: tokenizer.ggml.token_type arr[i32,102400] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 31: tokenizer.ggml.merges arr[str,99757] = ["Ġ Ġ", "Ġ t", "Ġ a", "i n", "h e... llama_model_loader: - kv 32: tokenizer.ggml.bos_token_id u32 = 100000 llama_model_loader: - kv 33: tokenizer.ggml.eos_token_id u32 = 100001 llama_model_loader: - kv 34: tokenizer.ggml.padding_token_id u32 = 100001 llama_model_loader: - kv 35: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 36: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 37: tokenizer.chat_template str = {% if not add_generation_prompt is de... llama_model_loader: - kv 38: general.quantization_version u32 = 2 llama_model_loader: - type f32: 300 tensors llama_model_loader: - type q8_0: 659 tensors llm_load_vocab: special tokens cache size = 2400 llm_load_vocab: token to piece cache size = 0.6661 MB llm_load_print_meta: format = GGUF V3 (latest) llm_load_print_meta: arch = deepseek2 llm_load_print_meta: vocab type = BPE llm_load_print_meta: n_vocab = 102400 llm_load_print_meta: n_merges = 99757 llm_load_print_meta: n_ctx_train = 163840 llm_load_print_meta: n_embd = 5120 llm_load_print_meta: n_head = 128 llm_load_print_meta: n_head_kv = 128 llm_load_print_meta: n_layer = 60 llm_load_print_meta: n_rot = 64 llm_load_print_meta: n_embd_head_k = 192 llm_load_print_meta: n_embd_head_v = 128 llm_load_print_meta: n_gqa = 1 llm_load_print_meta: n_embd_k_gqa = 24576 llm_load_print_meta: n_embd_v_gqa = 16384 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 = 12288 llm_load_print_meta: n_expert = 160 llm_load_print_meta: n_expert_used = 6 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 = yarn llm_load_print_meta: freq_base_train = 10000.0 llm_load_print_meta: freq_scale_train = 0.025 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 = 236B llm_load_print_meta: model ftype = Q8_0 llm_load_print_meta: model params = 235.74 B llm_load_print_meta: model size = 233.41 GiB (8.50 BPW) llm_load_print_meta: general.name = DeepSeek-Coder-V2-Instruct llm_load_print_meta: BOS token = 100000 '<｜begin▁of▁sentence｜>' llm_load_print_meta: EOS token = 100001 '<｜end▁of▁sentence｜>' llm_load_print_meta: PAD token = 100001 '<｜end▁of▁sentence｜>' llm_load_print_meta: LF token = 126 'Ä' llm_load_print_meta: n_layer_dense_lead = 1 llm_load_print_meta: n_lora_q = 1536 llm_load_print_meta: n_lora_kv = 512 llm_load_print_meta: n_ff_exp = 1536 llm_load_print_meta: n_expert_shared = 2 llm_load_print_meta: expert_weights_scale = 16.0 llm_load_print_meta: rope_yarn_log_mul = 0.1000 ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no ggml_cuda_init: CUDA_USE_TENSOR_CORES: yes 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 = 0.87 MiB llm_load_tensors: offloading 5 repeating layers to GPU llm_load_tensors: offloaded 5/61 layers to GPU llm_load_tensors: CPU buffer size = 218873.36 MiB llm_load_tensors: CUDA0 buffer size = 20135.96 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 = 0.025 llama_kv_cache_init: CUDA_Host KV buffer size = 2200.00 MiB llama_kv_cache_init: CUDA0 KV buffer size = 200.00 MiB llama_new_context_with_model: KV self size = 2400.00 MiB, K (f16): 1440.00 MiB, V (f16): 960.00 MiB llama_new_context_with_model: CUDA_Host output buffer size = 0.39 MiB llama_new_context_with_model: CUDA0 compute buffer size = 1422.00 MiB llama_new_context_with_model: CUDA_Host compute buffer size = 81.01 MiB llama_new_context_with_model: graph nodes = 4480 llama_new_context_with_model: graph splits = 990 system_info: n_threads = 32 / 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 205.752 ms compute_imatrix: computing over 139 chunks with batch_size 512 compute_imatrix: 182.74 seconds per pass - ETA 7 hours 3.33 minutes [1]5.3405,[2]3.6828,[3]3.5942,[4]3.9455,[5]3.8909,[6]3.6837,[7]3.8914,[8]3.7487,[9]4.1827, save_imatrix: stored collected data after 10 chunks in DeepSeek-Coder-V2-Instruct-IMat-GGUF/imatrix.dat [10]4.3749,[11]4.0626,[12]4.3081,[13]4.6410,[14]4.8878,[15]5.0294,[16]5.2433,[17]5.4254,[18]5.5530,[19]5.6628, save_imatrix: stored collected data after 20 chunks in DeepSeek-Coder-V2-Instruct-IMat-GGUF/imatrix.dat [20]5.4039,[21]5.4162,[22]5.3538,[23]5.3802,[24]5.3051,[25]5.4374,[26]5.3313,[27]5.4499,[28]5.3161,[29]5.0592, save_imatrix: stored collected data after 30 chunks in DeepSeek-Coder-V2-Instruct-IMat-GGUF/imatrix.dat [30]4.8366,[31]4.7855,[32]4.7328,[33]4.6131,[34]4.4173,[35]4.2519,[36]4.2044,[37]4.1553,[38]4.1683,[39]4.1118, save_imatrix: stored collected data after 40 chunks in DeepSeek-Coder-V2-Instruct-IMat-GGUF/imatrix.dat [40]4.0698,[41]4.0099,[42]3.9540,[43]3.9966,[44]4.0616,[45]4.1490,[46]4.1685,[47]4.0520,[48]3.9426,[49]3.8461, save_imatrix: stored collected data after 50 chunks in DeepSeek-Coder-V2-Instruct-IMat-GGUF/imatrix.dat [50]3.7519,[51]3.7617,[52]3.7385,[53]3.8232,[54]3.8956,[55]3.9832,[56]3.9474,[57]3.9662,[58]3.9917,[59]4.0548, save_imatrix: stored collected data after 60 chunks in DeepSeek-Coder-V2-Instruct-IMat-GGUF/imatrix.dat [60]4.1186,[61]4.1814,[62]4.2280,[63]4.2580,[64]4.2944,[65]4.3234,[66]4.3207,[67]4.3274,[68]4.3326,[69]4.3722, save_imatrix: stored collected data after 70 chunks in DeepSeek-Coder-V2-Instruct-IMat-GGUF/imatrix.dat [70]4.3903,[71]4.3980,[72]4.4266,[73]4.4352,[74]4.4371,[75]4.4443,[76]4.4506,[77]4.4644,[78]4.4901,[79]4.4807, save_imatrix: stored collected data after 80 chunks in DeepSeek-Coder-V2-Instruct-IMat-GGUF/imatrix.dat [80]4.5037,[81]4.5024,[82]4.5086,[83]4.5045,[84]4.5093,[85]4.5071,[86]4.5072,[87]4.4979,[88]4.5244,[89]4.5483, save_imatrix: stored collected data after 90 chunks in DeepSeek-Coder-V2-Instruct-IMat-GGUF/imatrix.dat [90]4.5559,[91]4.5866,[92]4.6173,[93]4.5939,[94]4.5951,[95]4.5810,[96]4.6027,[97]4.6223,[98]4.6212,[99]4.5856, save_imatrix: stored collected data after 100 chunks in DeepSeek-Coder-V2-Instruct-IMat-GGUF/imatrix.dat [100]4.5503,[101]4.5139,[102]4.4777,[103]4.4426,[104]4.4117,[105]4.3791,[106]4.3464,[107]4.3162,[108]4.2935,[109]4.3083, save_imatrix: stored collected data after 110 chunks in DeepSeek-Coder-V2-Instruct-IMat-GGUF/imatrix.dat [110]4.3381,[111]4.3819,[112]4.4259,[113]4.4635,[114]4.5322,[115]4.5670,[116]4.5901,[117]4.5989,[118]4.6225,[119]4.6201, save_imatrix: stored collected data after 120 chunks in DeepSeek-Coder-V2-Instruct-IMat-GGUF/imatrix.dat [120]4.5911,[121]4.5398,[122]4.4909,[123]4.5192,[124]4.5508,[125]4.5550,[126]4.5606,[127]4.5730,[128]4.5988,[129]4.6047, save_imatrix: stored collected data after 130 chunks in DeepSeek-Coder-V2-Instruct-IMat-GGUF/imatrix.dat [130]4.6171,[131]4.6390,[132]4.6376,[133]4.6276,[134]4.5805,[135]4.5325,[136]4.5336,[137]4.4897,[138]4.4510,[139]4.4089, save_imatrix: stored collected data after 139 chunks in DeepSeek-Coder-V2-Instruct-IMat-GGUF/imatrix.dat llama_print_timings: load time = 382279.94 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 = 29899992.31 ms / 71168 tokens ( 420.13 ms per token, 2.38 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 = 30105986.13 ms / 71169 tokens Final estimate: PPL = 4.4089 +/- 0.05371