llama_model_loader: loaded meta data with 35 key-value pairs and 363 tensors from Mistral-NeMo-Minitron-8B-Base-IMat-GGUF/Mistral-NeMo-Minitron-8B-Base.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.type str = model llama_model_loader: - kv 2: general.name str = Mistral NeMo Minitron 8B Base llama_model_loader: - kv 3: general.organization str = Nvidia llama_model_loader: - kv 4: general.finetune str = Base llama_model_loader: - kv 5: general.basename str = Mistral-NeMo-Minitron llama_model_loader: - kv 6: general.size_label str = 8B llama_model_loader: - kv 7: general.license str = other llama_model_loader: - kv 8: general.license.name str = nvidia-open-model-license llama_model_loader: - kv 9: general.license.link str = https://developer.download.nvidia.com... llama_model_loader: - kv 10: llama.block_count u32 = 40 llama_model_loader: - kv 11: llama.context_length u32 = 8192 llama_model_loader: - kv 12: llama.embedding_length u32 = 4096 llama_model_loader: - kv 13: llama.feed_forward_length u32 = 11520 llama_model_loader: - kv 14: llama.attention.head_count u32 = 32 llama_model_loader: - kv 15: llama.attention.head_count_kv u32 = 8 llama_model_loader: - kv 16: llama.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 17: llama.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 18: llama.attention.key_length u32 = 128 llama_model_loader: - kv 19: llama.attention.value_length u32 = 128 llama_model_loader: - kv 20: general.file_type u32 = 7 llama_model_loader: - kv 21: llama.vocab_size u32 = 131072 llama_model_loader: - kv 22: llama.rope.dimension_count u32 = 128 llama_model_loader: - kv 23: tokenizer.ggml.add_space_prefix bool = false llama_model_loader: - kv 24: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 25: tokenizer.ggml.pre str = tekken llama_model_loader: - kv 26: tokenizer.ggml.tokens arr[str,131072] = ["", "", "", "[INST]", "[... llama_model_loader: - kv 27: tokenizer.ggml.token_type arr[i32,131072] = [3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, ... llama_model_loader: - kv 28: tokenizer.ggml.merges arr[str,269443] = ["Ä  Ä ", "Ä  t", "e r", "i n", "Ä  Ä... llama_model_loader: - kv 29: tokenizer.ggml.bos_token_id u32 = 1 llama_model_loader: - kv 30: tokenizer.ggml.eos_token_id u32 = 2 llama_model_loader: - kv 31: tokenizer.ggml.unknown_token_id u32 = 0 llama_model_loader: - kv 32: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 33: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 34: general.quantization_version u32 = 2 llama_model_loader: - type f32: 81 tensors llama_model_loader: - type q8_0: 282 tensors llm_load_vocab: special tokens cache size = 1000 llm_load_vocab: token to piece cache size = 0.8498 MB llm_load_print_meta: format = GGUF V3 (latest) llm_load_print_meta: arch = llama llm_load_print_meta: vocab type = BPE llm_load_print_meta: n_vocab = 131072 llm_load_print_meta: n_merges = 269443 llm_load_print_meta: vocab_only = 0 llm_load_print_meta: n_ctx_train = 8192 llm_load_print_meta: n_embd = 4096 llm_load_print_meta: n_layer = 40 llm_load_print_meta: n_head = 32 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 = 4 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-05 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 = 11520 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 = 1000000.0 llm_load_print_meta: freq_scale_train = 1 llm_load_print_meta: n_ctx_orig_yarn = 8192 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 = 13B llm_load_print_meta: model ftype = Q8_0 llm_load_print_meta: model params = 8.41 B llm_load_print_meta: model size = 8.33 GiB (8.50 BPW) llm_load_print_meta: general.name = Mistral NeMo Minitron 8B Base llm_load_print_meta: BOS token = 1 '' llm_load_print_meta: EOS token = 2 '' llm_load_print_meta: UNK token = 0 '' llm_load_print_meta: LF token = 1196 'Ä' llm_load_print_meta: max token length = 150 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 = 0.34 MiB llm_load_tensors: offloading 40 repeating layers to GPU llm_load_tensors: offloading non-repeating layers to GPU llm_load_tensors: offloaded 41/41 layers to GPU llm_load_tensors: CPU buffer size = 544.00 MiB llm_load_tensors: CUDA0 buffer size = 7982.78 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: CUDA0 KV buffer size = 80.00 MiB llama_new_context_with_model: KV self size = 80.00 MiB, K (f16): 40.00 MiB, V (f16): 40.00 MiB llama_new_context_with_model: CUDA_Host output buffer size = 0.50 MiB llama_new_context_with_model: CUDA0 compute buffer size = 264.00 MiB llama_new_context_with_model: CUDA_Host compute buffer size = 9.01 MiB llama_new_context_with_model: graph nodes = 1286 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 108.019 ms compute_imatrix: computing over 128 chunks with batch_size 512 compute_imatrix: 0.81 seconds per pass - ETA 1.72 minutes [1]5.0917,[2]3.8243,[3]3.5221,[4]4.3283,[5]4.4019,[6]3.8244,[7]4.1977,[8]4.4898,[9]4.7123, save_imatrix: stored collected data after 10 chunks in Mistral-NeMo-Minitron-8B-Base-IMat-GGUF/imatrix.dat [10]4.2822,[11]4.5663,[12]4.8948,[13]5.2375,[14]5.4594,[15]5.7700,[16]5.9816,[17]6.1292,[18]6.3001,[19]6.0685, save_imatrix: stored collected data after 20 chunks in Mistral-NeMo-Minitron-8B-Base-IMat-GGUF/imatrix.dat [20]6.1563,[21]6.3118,[22]6.2370,[23]6.3808,[24]6.3845,[25]6.5794,[26]6.4172,[27]6.0985,[28]6.1277,[29]6.0738, save_imatrix: stored collected data after 30 chunks in Mistral-NeMo-Minitron-8B-Base-IMat-GGUF/imatrix.dat [30]6.0397,[31]5.7730,[32]5.6212,[33]5.5789,[34]5.5094,[35]5.4731,[36]5.6957,[37]5.6947,[38]5.7691,[39]5.8936, save_imatrix: stored collected data after 40 chunks in Mistral-NeMo-Minitron-8B-Base-IMat-GGUF/imatrix.dat [40]6.0177,[41]6.1334,[42]6.2400,[43]6.4518,[44]6.6405,[45]6.7468,[46]6.6502,[47]6.6859,[48]6.8168,[49]6.9292, save_imatrix: stored collected data after 50 chunks in Mistral-NeMo-Minitron-8B-Base-IMat-GGUF/imatrix.dat [50]6.7899,[51]6.7992,[52]6.8174,[53]6.8827,[54]6.9915,[55]7.0901,[56]7.1331,[57]7.1425,[58]7.1123,[59]7.0535, save_imatrix: stored collected data after 60 chunks in Mistral-NeMo-Minitron-8B-Base-IMat-GGUF/imatrix.dat [60]7.0041,[61]6.9069,[62]6.8566,[63]6.9065,[64]6.9412,[65]6.8610,[66]6.8269,[67]6.8264,[68]6.7904,[69]6.7598, save_imatrix: stored collected data after 70 chunks in Mistral-NeMo-Minitron-8B-Base-IMat-GGUF/imatrix.dat [70]6.7628,[71]6.7871,[72]6.7442,[73]6.7638,[74]6.7438,[75]6.7203,[76]6.7261,[77]6.6959,[78]6.6498,[79]6.5994, save_imatrix: stored collected data after 80 chunks in Mistral-NeMo-Minitron-8B-Base-IMat-GGUF/imatrix.dat [80]6.6226,[81]6.6484,[82]6.6333,[83]6.6263,[84]6.6079,[85]6.6463,[86]6.5846,[87]6.5644,[88]6.5804,[89]6.5922, save_imatrix: stored collected data after 90 chunks in Mistral-NeMo-Minitron-8B-Base-IMat-GGUF/imatrix.dat [90]6.5901,[91]6.5320,[92]6.4827,[93]6.4232,[94]6.3606,[95]6.3125,[96]6.2706,[97]6.2153,[98]6.1597,[99]6.1305, save_imatrix: stored collected data after 100 chunks in Mistral-NeMo-Minitron-8B-Base-IMat-GGUF/imatrix.dat [100]6.1450,[101]6.2110,[102]6.2667,[103]6.3205,[104]6.3722,[105]6.4656,[106]6.4593,[107]6.4949,[108]6.4325,[109]6.4323, save_imatrix: stored collected data after 110 chunks in Mistral-NeMo-Minitron-8B-Base-IMat-GGUF/imatrix.dat [110]6.3983,[111]6.3213,[112]6.2552,[113]6.2476,[114]6.2984,[115]6.2985,[116]6.2961,[117]6.3105,[118]6.3396,[119]6.3363, save_imatrix: stored collected data after 120 chunks in Mistral-NeMo-Minitron-8B-Base-IMat-GGUF/imatrix.dat [120]6.3302,[121]6.3419,[122]6.3261,[123]6.3507,[124]6.3745,[125]6.3947,[126]6.4361,[127]6.4525,[128]6.4737, save_imatrix: stored collected data after 128 chunks in Mistral-NeMo-Minitron-8B-Base-IMat-GGUF/imatrix.dat llama_print_timings: load time = 2332.08 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 = 80639.92 ms / 65536 tokens ( 1.23 ms per token, 812.70 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 = 83049.27 ms / 65537 tokens Final estimate: PPL = 6.4737 +/- 0.08410