llama_model_loader: loaded meta data with 24 key-value pairs and 363 tensors from llm-compiler-13b-IMat-GGUF/llm-compiler-13b.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 = llm-compiler-13b llama_model_loader: - kv 2: llama.block_count u32 = 40 llama_model_loader: - kv 3: llama.context_length u32 = 16384 llama_model_loader: - kv 4: llama.embedding_length u32 = 5120 llama_model_loader: - kv 5: llama.feed_forward_length u32 = 13824 llama_model_loader: - kv 6: llama.attention.head_count u32 = 40 llama_model_loader: - kv 7: llama.attention.head_count_kv u32 = 40 llama_model_loader: - kv 8: llama.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 9: llama.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 10: general.file_type u32 = 7 llama_model_loader: - kv 11: llama.vocab_size u32 = 32000 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,32000] = ["", "~~", "~~", "<0x00>", "<... llama_model_loader: - kv 16: tokenizer.ggml.scores arr[f32,32000] = [0.000000, 0.000000, 0.000000, 0.0000... llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,32000] = [2, 3, 3, 6, 6, 6, 6, 6, 6, 6, 6, 6, ... 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.add_bos_token bool = true llama_model_loader: - kv 22: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 23: 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 = 259 llm_load_vocab: token to piece cache size = 0.1684 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 = 32000 llm_load_print_meta: n_merges = 0 llm_load_print_meta: n_ctx_train = 16384 llm_load_print_meta: n_embd = 5120 llm_load_print_meta: n_head = 40 llm_load_print_meta: n_head_kv = 40 llm_load_print_meta: n_layer = 40 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 = 1 llm_load_print_meta: n_embd_k_gqa = 5120 llm_load_print_meta: n_embd_v_gqa = 5120 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 = 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 = 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 = 16384 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 = 13B llm_load_print_meta: model ftype = Q8_0 llm_load_print_meta: model params = 13.02 B llm_load_print_meta: model size = 12.88 GiB (8.50 BPW) llm_load_print_meta: general.name = llm-compiler-13b 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 = 13 '<0x0A>' 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 = 0.34 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 = 166.02 MiB llm_load_tensors: CUDA0 buffer size = 13023.85 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 = 400.00 MiB llama_new_context_with_model: KV self size = 400.00 MiB, K (f16): 200.00 MiB, V (f16): 200.00 MiB llama_new_context_with_model: CUDA_Host output buffer size = 0.12 MiB llama_new_context_with_model: CUDA0 compute buffer size = 85.00 MiB llama_new_context_with_model: CUDA_Host compute buffer size = 11.01 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 92.657 ms compute_imatrix: computing over 151 chunks with batch_size 512 compute_imatrix: 0.84 seconds per pass - ETA 2.12 minutes [1]6.5102,[2]4.8278,[3]4.8906,[4]5.8275,[5]6.5980,[6]6.6741,[7]6.1521,[8]6.6585,[9]6.8038, save_imatrix: stored collected data after 10 chunks in llm-compiler-13b-IMat-GGUF/imatrix.dat [10]7.2034,[11]7.0897,[12]6.3110,[13]6.1925,[14]6.4733,[15]6.9117,[16]7.0055,[17]7.3541,[18]7.5604,[19]7.7247, save_imatrix: stored collected data after 20 chunks in llm-compiler-13b-IMat-GGUF/imatrix.dat [20]7.8130,[21]8.0770,[22]7.6913,[23]7.3320,[24]7.3916,[25]7.4575,[26]7.4139,[27]7.2614,[28]7.3498,[29]7.4908, save_imatrix: stored collected data after 30 chunks in llm-compiler-13b-IMat-GGUF/imatrix.dat [30]7.6196,[31]7.6048,[32]7.7227,[33]7.8117,[34]8.0091,[35]8.0498,[36]7.9688,[37]7.6228,[38]7.3958,[39]7.3474, save_imatrix: stored collected data after 40 chunks in llm-compiler-13b-IMat-GGUF/imatrix.dat [40]7.2737,[41]7.1858,[42]7.1084,[43]6.9552,[44]6.8893,[45]6.7978,[46]6.7678,[47]6.7896,[48]6.8411,[49]6.9229, save_imatrix: stored collected data after 50 chunks in llm-compiler-13b-IMat-GGUF/imatrix.dat [50]6.9341,[51]7.0692,[52]7.1922,[53]7.3284,[54]7.4554,[55]7.5083,[56]7.4600,[57]7.3916,[58]7.4462,[59]7.5115, save_imatrix: stored collected data after 60 chunks in llm-compiler-13b-IMat-GGUF/imatrix.dat [60]7.5997,[61]7.5003,[62]7.5256,[63]7.5999,[64]7.6843,[65]7.7434,[66]7.7791,[67]7.8386,[68]7.8891,[69]7.8758, save_imatrix: stored collected data after 70 chunks in llm-compiler-13b-IMat-GGUF/imatrix.dat [70]7.8826,[71]7.8861,[72]7.8123,[73]7.7649,[74]7.7193,[75]7.7113,[76]7.7044,[77]7.7125,[78]7.6880,[79]7.7022, save_imatrix: stored collected data after 80 chunks in llm-compiler-13b-IMat-GGUF/imatrix.dat [80]7.7223,[81]7.7006,[82]7.6871,[83]7.6427,[84]7.6579,[85]7.6692,[86]7.6653,[87]7.6858,[88]7.6954,[89]7.6857, save_imatrix: stored collected data after 90 chunks in llm-compiler-13b-IMat-GGUF/imatrix.dat [90]7.6656,[91]7.6720,[92]7.6310,[93]7.6197,[94]7.5937,[95]7.5636,[96]7.5907,[97]7.5873,[98]7.5975,[99]7.5835, save_imatrix: stored collected data after 100 chunks in llm-compiler-13b-IMat-GGUF/imatrix.dat [100]7.5752,[101]7.5954,[102]7.5642,[103]7.5355,[104]7.5224,[105]7.5463,[106]7.5487,[107]7.5597,[108]7.5833,[109]7.5193, save_imatrix: stored collected data after 110 chunks in llm-compiler-13b-IMat-GGUF/imatrix.dat [110]7.4666,[111]7.4126,[112]7.3521,[113]7.2913,[114]7.2362,[115]7.1853,[116]7.1342,[117]7.0951,[118]7.1096,[119]7.1196, save_imatrix: stored collected data after 120 chunks in llm-compiler-13b-IMat-GGUF/imatrix.dat [120]7.1684,[121]7.2164,[122]7.2682,[123]7.3216,[124]7.4055,[125]7.4902,[126]7.5035,[127]7.5218,[128]7.4561,[129]7.4576, save_imatrix: stored collected data after 130 chunks in llm-compiler-13b-IMat-GGUF/imatrix.dat [130]7.4422,[131]7.4365,[132]7.4118,[133]7.4061,[134]7.4185,[135]7.4420,[136]7.4324,[137]7.4280,[138]7.4363,[139]7.4519, save_imatrix: stored collected data after 140 chunks in llm-compiler-13b-IMat-GGUF/imatrix.dat [140]7.4695,[141]7.4749,[142]7.4733,[143]7.4593,[144]7.4382,[145]7.4538,[146]7.4791,[147]7.5098,[148]7.5349,[149]7.5722, save_imatrix: stored collected data after 150 chunks in llm-compiler-13b-IMat-GGUF/imatrix.dat [150]7.6062,[151]7.6399, save_imatrix: stored collected data after 151 chunks in llm-compiler-13b-IMat-GGUF/imatrix.dat llama_print_timings: load time = 10260.16 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 = 119297.93 ms / 77312 tokens ( 1.54 ms per token, 648.06 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 = 129182.97 ms / 77313 tokens Final estimate: PPL = 7.6399 +/- 0.09694