llama_model_loader: loaded meta data with 24 key-value pairs and 291 tensors from llm-compiler-7b-IMat-GGUF/llm-compiler-7b.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-7b llama_model_loader: - kv 2: llama.block_count u32 = 32 llama_model_loader: - kv 3: llama.context_length u32 = 16384 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 = 32 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: 65 tensors llama_model_loader: - type q8_0: 226 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 = 4096 llm_load_print_meta: n_head = 32 llm_load_print_meta: n_head_kv = 32 llm_load_print_meta: n_layer = 32 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 = 4096 llm_load_print_meta: n_embd_v_gqa = 4096 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 = 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 = 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 = 7B llm_load_print_meta: model ftype = Q8_0 llm_load_print_meta: model params = 6.74 B llm_load_print_meta: model size = 6.67 GiB (8.50 BPW) llm_load_print_meta: general.name = llm-compiler-7b 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 = 0.27 MiB llm_load_tensors: offloading 32 repeating layers to GPU llm_load_tensors: offloading non-repeating layers to GPU llm_load_tensors: offloaded 33/33 layers to GPU llm_load_tensors: CPU buffer size = 132.81 MiB llm_load_tensors: CUDA0 buffer size = 6695.84 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 = 256.00 MiB llama_new_context_with_model: KV self size = 256.00 MiB, K (f16): 128.00 MiB, V (f16): 128.00 MiB llama_new_context_with_model: CUDA_Host output buffer size = 0.12 MiB llama_new_context_with_model: CUDA0 compute buffer size = 70.50 MiB llama_new_context_with_model: CUDA_Host compute buffer size = 9.01 MiB llama_new_context_with_model: graph nodes = 1030 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 98.047 ms compute_imatrix: computing over 151 chunks with batch_size 512 compute_imatrix: 0.59 seconds per pass - ETA 1.47 minutes [1]6.6836,[2]5.1025,[3]5.0990,[4]6.0509,[5]6.8286,[6]6.9178,[7]6.4531,[8]7.0779,[9]7.2172, save_imatrix: stored collected data after 10 chunks in llm-compiler-7b-IMat-GGUF/imatrix.dat [10]7.6626,[11]7.6180,[12]6.7697,[13]6.6368,[14]6.9780,[15]7.4385,[16]7.5302,[17]7.8579,[18]8.0717,[19]8.2099, save_imatrix: stored collected data after 20 chunks in llm-compiler-7b-IMat-GGUF/imatrix.dat [20]8.3085,[21]8.5583,[22]8.1441,[23]7.7425,[24]7.8041,[25]7.8916,[26]7.8419,[27]7.6914,[28]7.8043,[29]7.9600, save_imatrix: stored collected data after 30 chunks in llm-compiler-7b-IMat-GGUF/imatrix.dat [30]8.0985,[31]8.1050,[32]8.2265,[33]8.3044,[34]8.5184,[35]8.5777,[36]8.4893,[37]8.1221,[38]7.8718,[39]7.7932, save_imatrix: stored collected data after 40 chunks in llm-compiler-7b-IMat-GGUF/imatrix.dat [40]7.7251,[41]7.6244,[42]7.5451,[43]7.3857,[44]7.2989,[45]7.2049,[46]7.1984,[47]7.2254,[48]7.2786,[49]7.3786, save_imatrix: stored collected data after 50 chunks in llm-compiler-7b-IMat-GGUF/imatrix.dat [50]7.3834,[51]7.5855,[52]7.7687,[53]7.9564,[54]8.1367,[55]8.2275,[56]8.1763,[57]8.1040,[58]8.1407,[59]8.2010, save_imatrix: stored collected data after 60 chunks in llm-compiler-7b-IMat-GGUF/imatrix.dat [60]8.2955,[61]8.1878,[62]8.2123,[63]8.2870,[64]8.3726,[65]8.4340,[66]8.4646,[67]8.5276,[68]8.5576,[69]8.5352, save_imatrix: stored collected data after 70 chunks in llm-compiler-7b-IMat-GGUF/imatrix.dat [70]8.5404,[71]8.5466,[72]8.4782,[73]8.4271,[74]8.3748,[75]8.3548,[76]8.3520,[77]8.3590,[78]8.3342,[79]8.3392, save_imatrix: stored collected data after 80 chunks in llm-compiler-7b-IMat-GGUF/imatrix.dat [80]8.3615,[81]8.3437,[82]8.3309,[83]8.2756,[84]8.2916,[85]8.2987,[86]8.2900,[87]8.3182,[88]8.3145,[89]8.3015, save_imatrix: stored collected data after 90 chunks in llm-compiler-7b-IMat-GGUF/imatrix.dat [90]8.2822,[91]8.2930,[92]8.2448,[93]8.2390,[94]8.2104,[95]8.1747,[96]8.1994,[97]8.1919,[98]8.2019,[99]8.1808, save_imatrix: stored collected data after 100 chunks in llm-compiler-7b-IMat-GGUF/imatrix.dat [100]8.1697,[101]8.1935,[102]8.1551,[103]8.1190,[104]8.1010,[105]8.1276,[106]8.1249,[107]8.1380,[108]8.1674,[109]8.0909, save_imatrix: stored collected data after 110 chunks in llm-compiler-7b-IMat-GGUF/imatrix.dat [110]8.0305,[111]7.9669,[112]7.9005,[113]7.8343,[114]7.7723,[115]7.7137,[116]7.6576,[117]7.6158,[118]7.6329,[119]7.6453, save_imatrix: stored collected data after 120 chunks in llm-compiler-7b-IMat-GGUF/imatrix.dat [120]7.6954,[121]7.7506,[122]7.8016,[123]7.8576,[124]7.9541,[125]8.0482,[126]8.0601,[127]8.0794,[128]8.0077,[129]8.0114, save_imatrix: stored collected data after 130 chunks in llm-compiler-7b-IMat-GGUF/imatrix.dat [130]7.9939,[131]7.9857,[132]7.9606,[133]7.9555,[134]7.9690,[135]7.9940,[136]7.9830,[137]7.9785,[138]7.9913,[139]8.0043, save_imatrix: stored collected data after 140 chunks in llm-compiler-7b-IMat-GGUF/imatrix.dat [140]8.0231,[141]8.0263,[142]8.0282,[143]8.0130,[144]7.9915,[145]8.0111,[146]8.0400,[147]8.0742,[148]8.1063,[149]8.1451, save_imatrix: stored collected data after 150 chunks in llm-compiler-7b-IMat-GGUF/imatrix.dat [150]8.1810,[151]8.2203, save_imatrix: stored collected data after 151 chunks in llm-compiler-7b-IMat-GGUF/imatrix.dat llama_print_timings: load time = 1857.95 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 = 81192.92 ms / 77312 tokens ( 1.05 ms per token, 952.20 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 = 82973.02 ms / 77313 tokens Final estimate: PPL = 8.2203 +/- 0.10580