llama_model_loader: loaded meta data with 21 key-value pairs and 963 tensors from Qwen2-72B-Instruct-IMat-GGUF/Qwen2-72B-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.name str = Qwen2-72B-Instruct llama_model_loader: - kv 2: qwen2.block_count u32 = 80 llama_model_loader: - kv 3: qwen2.context_length u32 = 32768 llama_model_loader: - kv 4: qwen2.embedding_length u32 = 8192 llama_model_loader: - kv 5: qwen2.feed_forward_length u32 = 29568 llama_model_loader: - kv 6: qwen2.attention.head_count u32 = 64 llama_model_loader: - kv 7: qwen2.attention.head_count_kv u32 = 8 llama_model_loader: - kv 8: qwen2.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 9: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 10: general.file_type u32 = 7 llama_model_loader: - kv 11: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 12: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 13: tokenizer.ggml.tokens arr[str,152064] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 14: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 15: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 16: tokenizer.ggml.eos_token_id u32 = 151645 llama_model_loader: - kv 17: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 18: tokenizer.ggml.bos_token_id u32 = 151643 llama_model_loader: - kv 19: tokenizer.chat_template str = {% for message in messages %}{% if lo... llama_model_loader: - kv 20: general.quantization_version u32 = 2 llama_model_loader: - type f32: 401 tensors llama_model_loader: - type q8_0: 562 tensors llm_load_vocab: special tokens cache size = 421 llm_load_vocab: token to piece cache size = 0.9352 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: n_ctx_train = 32768 llm_load_print_meta: n_embd = 8192 llm_load_print_meta: n_head = 64 llm_load_print_meta: n_head_kv = 8 llm_load_print_meta: n_layer = 80 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 = 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 = 29568 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: model type = 70B llm_load_print_meta: model ftype = Q8_0 llm_load_print_meta: model params = 72.71 B llm_load_print_meta: model size = 71.95 GiB (8.50 BPW) llm_load_print_meta: general.name = Qwen2-72B-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|>' 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 = 0.92 MiB llm_load_tensors: offloading 20 repeating layers to GPU llm_load_tensors: offloaded 20/81 layers to GPU llm_load_tensors: CPU buffer size = 73677.66 MiB llm_load_tensors: CUDA0 buffer size = 17788.29 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 = 120.00 MiB llama_kv_cache_init: CUDA0 KV buffer size = 40.00 MiB llama_new_context_with_model: KV self size = 160.00 MiB, K (f16): 80.00 MiB, V (f16): 80.00 MiB llama_new_context_with_model: CUDA_Host output buffer size = 0.58 MiB llama_new_context_with_model: CUDA0 compute buffer size = 1575.25 MiB llama_new_context_with_model: CUDA_Host compute buffer size = 17.01 MiB llama_new_context_with_model: graph nodes = 2806 llama_new_context_with_model: graph splits = 844 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 136.611 ms compute_imatrix: computing over 128 chunks with batch_size 512 compute_imatrix: 27.81 seconds per pass - ETA 59.32 minutes [1]4.0667,[2]2.9968,[3]2.7877,[4]3.0396,[5]3.0193,[6]2.7974,[7]2.9432,[8]3.0042,[9]3.3837, save_imatrix: stored collected data after 10 chunks in Qwen2-72B-Instruct-IMat-GGUF/imatrix.dat [10]3.3551,[11]3.3012,[12]3.6225,[13]3.9980,[14]4.2226,[15]4.5451,[16]4.8045,[17]4.9530,[18]5.1710,[19]5.0171, save_imatrix: stored collected data after 20 chunks in Qwen2-72B-Instruct-IMat-GGUF/imatrix.dat [20]5.1555,[21]5.1434,[22]5.1624,[23]5.1016,[24]5.2596,[25]5.3729,[26]5.2817,[27]5.3150,[28]5.3345,[29]5.4403, save_imatrix: stored collected data after 30 chunks in Qwen2-72B-Instruct-IMat-GGUF/imatrix.dat [30]5.3995,[31]5.2560,[32]5.0742,[33]4.9801,[34]4.9146,[35]4.8823,[36]4.8909,[37]4.9080,[38]4.9352,[39]4.9022, save_imatrix: stored collected data after 40 chunks in Qwen2-72B-Instruct-IMat-GGUF/imatrix.dat [40]5.0046,[41]5.0428,[42]5.1636,[43]5.0551,[44]5.1610,[45]5.2157,[46]5.2933,[47]5.2076,[48]5.2499,[49]5.3324, save_imatrix: stored collected data after 50 chunks in Qwen2-72B-Instruct-IMat-GGUF/imatrix.dat [50]5.3917,[51]5.3295,[52]5.4023,[53]5.4976,[54]5.5736,[55]5.6064,[56]5.6562,[57]5.6968,[58]5.7409,[59]5.7721, save_imatrix: stored collected data after 60 chunks in Qwen2-72B-Instruct-IMat-GGUF/imatrix.dat [60]5.7922,[61]5.7742,[62]5.7481,[63]5.7837,[64]5.8379,[65]5.8041,[66]5.8132,[67]5.8367,[68]5.7894,[69]5.7512, save_imatrix: stored collected data after 70 chunks in Qwen2-72B-Instruct-IMat-GGUF/imatrix.dat [70]5.7483,[71]5.7360,[72]5.7379,[73]5.7505,[74]5.7101,[75]5.6809,[76]5.6496,[77]5.6416,[78]5.6378,[79]5.6251, save_imatrix: stored collected data after 80 chunks in Qwen2-72B-Instruct-IMat-GGUF/imatrix.dat [80]5.5726,[81]5.6016,[82]5.6046,[83]5.5769,[84]5.5958,[85]5.6151,[86]5.5989,[87]5.5806,[88]5.5546,[89]5.5692, save_imatrix: stored collected data after 90 chunks in Qwen2-72B-Instruct-IMat-GGUF/imatrix.dat [90]5.5935,[91]5.5874,[92]5.5595,[93]5.5326,[94]5.4937,[95]5.4613,[96]5.4304,[97]5.3971,[98]5.3679,[99]5.3492, save_imatrix: stored collected data after 100 chunks in Qwen2-72B-Instruct-IMat-GGUF/imatrix.dat [100]5.3608,[101]5.3831,[102]5.4434,[103]5.5090,[104]5.5541,[105]5.6454,[106]5.6999,[107]5.7232,[108]5.7158,[109]5.7137, save_imatrix: stored collected data after 110 chunks in Qwen2-72B-Instruct-IMat-GGUF/imatrix.dat [110]5.7052,[111]5.6696,[112]5.6187,[113]5.5809,[114]5.6217,[115]5.6262,[116]5.6315,[117]5.6514,[118]5.6844,[119]5.6896, save_imatrix: stored collected data after 120 chunks in Qwen2-72B-Instruct-IMat-GGUF/imatrix.dat [120]5.6892,[121]5.7048,[122]5.6845,[123]5.7023,[124]5.7156,[125]5.7275,[126]5.7575,[127]5.7766,[128]5.8007, save_imatrix: stored collected data after 128 chunks in Qwen2-72B-Instruct-IMat-GGUF/imatrix.dat llama_print_timings: load time = 74513.74 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 = 1161436.60 ms / 65536 tokens ( 17.72 ms per token, 56.43 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 = 1210188.40 ms / 65537 tokens Final estimate: PPL = 5.8007 +/- 0.07472