main: build = 3010 (95f84d5c) main: built with cc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0 for x86_64-linux-gnu main: seed = 1716905720 llama_model_loader: loaded meta data with 27 key-value pairs and 291 tensors from AutoCoder_S_6.7B-IMat-GGUF/AutoCoder_S_6.7B.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 = AutoCoder_S_6.7B 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 = 100000.000000 llama_model_loader: - kv 9: llama.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 10: general.file_type u32 = 0 llama_model_loader: - kv 11: llama.vocab_size u32 = 32256 llama_model_loader: - kv 12: llama.rope.dimension_count u32 = 128 llama_model_loader: - kv 13: llama.rope.scaling.type str = linear llama_model_loader: - kv 14: llama.rope.scaling.factor f32 = 4.000000 llama_model_loader: - kv 15: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 16: tokenizer.ggml.pre str = deepseek-coder llama_model_loader: - kv 17: tokenizer.ggml.tokens arr[str,32256] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 18: tokenizer.ggml.token_type arr[i32,32256] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 19: tokenizer.ggml.merges arr[str,31757] = ["Ġ Ġ", "Ġ t", "Ġ a", "i n", "h e... llama_model_loader: - kv 20: tokenizer.ggml.bos_token_id u32 = 32013 llama_model_loader: - kv 21: tokenizer.ggml.eos_token_id u32 = 32021 llama_model_loader: - kv 22: tokenizer.ggml.padding_token_id u32 = 32014 llama_model_loader: - kv 23: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 24: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 25: tokenizer.chat_template str = {% if messages[0]['role'] == 'system'... llama_model_loader: - kv 26: general.quantization_version u32 = 2 llama_model_loader: - type f32: 291 tensors llm_load_vocab: mismatch in special tokens definition ( 243/32256 vs 256/32256 ). 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 = 32256 llm_load_print_meta: n_merges = 31757 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-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 = 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 = 100000.0 llm_load_print_meta: freq_scale_train = 0.25 llm_load_print_meta: n_yarn_orig_ctx = 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 = all F32 llm_load_print_meta: model params = 6.74 B llm_load_print_meta: model size = 25.11 GiB (32.00 BPW) llm_load_print_meta: general.name = AutoCoder_S_6.7B llm_load_print_meta: BOS token = 32013 '<｜begin▁of▁sentence｜>' llm_load_print_meta: EOS token = 32021 '<|EOT|>' llm_load_print_meta: PAD token = 32014 '<｜end▁of▁sentence｜>' llm_load_print_meta: LF token = 126 'Ä' 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 = 0.30 MiB llm_load_tensors: offloading 29 repeating layers to GPU llm_load_tensors: offloaded 29/33 layers to GPU llm_load_tensors: CPU buffer size = 25713.02 MiB llm_load_tensors: CUDA0 buffer size = 22388.91 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 = 100000.0 llama_new_context_with_model: freq_scale = 0.25 llama_kv_cache_init: CUDA_Host KV buffer size = 24.00 MiB llama_kv_cache_init: CUDA0 KV buffer size = 232.00 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 = 575.00 MiB llama_new_context_with_model: CUDA_Host compute buffer size = 17.01 MiB llama_new_context_with_model: graph nodes = 1030 llama_new_context_with_model: graph splits = 37 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 394.173 ms compute_imatrix: computing over 236 chunks with batch_size 512 compute_imatrix: 1.41 seconds per pass - ETA 5.53 minutes [1]6.9711,[2]5.6324,[3]5.7695,[4]6.9482,[5]7.1003,[6]6.8935,[7]6.0051,[8]6.8299,[9]6.5963, save_imatrix: stored collected data after 10 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [10]7.4973,[11]7.8150,[12]7.6111,[13]8.2735,[14]7.5730,[15]8.4049,[16]8.5410,[17]8.9150,[18]9.0491,[19]9.3996, save_imatrix: stored collected data after 20 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [20]9.1504,[21]9.4010,[22]9.2047,[23]8.7323,[24]8.8343,[25]8.2072,[26]7.7374,[27]7.4032,[28]7.2794,[29]7.3243, save_imatrix: stored collected data after 30 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [30]7.4102,[31]7.5618,[32]7.7531,[33]8.0081,[34]7.8609,[35]7.4665,[36]7.1310,[37]7.0749,[38]7.0595,[39]7.0472, save_imatrix: stored collected data after 40 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [40]7.0351,[41]7.1613,[42]7.3376,[43]7.4777,[44]7.7039,[45]7.6982,[46]7.8623,[47]8.0818,[48]8.2927,[49]8.5157, save_imatrix: stored collected data after 50 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [50]8.6643,[51]8.5805,[52]8.4288,[53]8.2782,[54]8.1212,[55]8.2830,[56]8.3925,[57]8.4649,[58]8.6312,[59]8.6761, save_imatrix: stored collected data after 60 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [60]8.8491,[61]9.0009,[62]9.1895,[63]9.3434,[64]9.4736,[65]9.5979,[66]9.6884,[67]9.8557,[68]9.9761,[69]10.0247, save_imatrix: stored collected data after 70 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [70]10.0730,[71]9.9550,[72]9.9019,[73]9.8934,[74]9.8477,[75]9.8404,[76]9.8001,[77]9.7517,[78]9.6413,[79]9.5902, save_imatrix: stored collected data after 80 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [80]9.5973,[81]9.5626,[82]9.6427,[83]9.7263,[84]9.8151,[85]9.6581,[86]9.6787,[87]9.6081,[88]9.6448,[89]9.7148, save_imatrix: stored collected data after 90 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [90]9.7689,[91]9.8819,[92]9.9310,[93]10.0121,[94]10.0795,[95]10.0701,[96]10.0096,[97]10.0003,[98]10.0176,[99]10.0750, save_imatrix: stored collected data after 100 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [100]10.1176,[101]10.1105,[102]10.1082,[103]10.0825,[104]10.0619,[105]10.0554,[106]10.0109,[107]9.9972,[108]9.9992,[109]9.9618, save_imatrix: stored collected data after 110 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [110]9.9424,[111]9.9028,[112]9.9015,[113]9.8885,[114]9.8615,[115]9.8325,[116]9.8164,[117]9.8120,[118]9.7936,[119]9.7131, save_imatrix: stored collected data after 120 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [120]9.7549,[121]9.7954,[122]9.8028,[123]9.7688,[124]9.7939,[125]9.8058,[126]9.7924,[127]9.7033,[128]9.7055,[129]9.7133, save_imatrix: stored collected data after 130 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [130]9.6597,[131]9.6724,[132]9.5981,[133]9.5206,[134]9.4413,[135]9.3635,[136]9.2893,[137]9.2113,[138]9.1416,[139]9.0679, save_imatrix: stored collected data after 140 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [140]9.0140,[141]8.9418,[142]8.8786,[143]8.8081,[144]8.7210,[145]8.6630,[146]8.6062,[147]8.5428,[148]8.4765,[149]8.4173, save_imatrix: stored collected data after 150 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [150]8.3599,[151]8.2933,[152]8.2349,[153]8.1789,[154]8.1178,[155]8.0694,[156]8.0121,[157]7.9777,[158]7.9050,[159]7.8430, save_imatrix: stored collected data after 160 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [160]7.8331,[161]7.8808,[162]7.9044,[163]7.9540,[164]8.0023,[165]7.9800,[166]8.0035,[167]7.9998,[168]7.9850,[169]7.9934, save_imatrix: stored collected data after 170 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [170]7.9962,[171]8.0055,[172]7.9904,[173]8.0206,[174]8.0128,[175]8.0323,[176]8.0310,[177]8.0447,[178]8.0503,[179]8.0647, save_imatrix: stored collected data after 180 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [180]8.0664,[181]8.0854,[182]8.1037,[183]8.1084,[184]8.1318,[185]8.1649,[186]8.2033,[187]8.2198,[188]8.2490,[189]8.2654, save_imatrix: stored collected data after 190 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [190]8.2910,[191]8.3175,[192]8.3504,[193]8.3763,[194]8.3824,[195]8.4359,[196]8.4523,[197]8.4449,[198]8.5025,[199]8.5595, save_imatrix: stored collected data after 200 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [200]8.6122,[201]8.6825,[202]8.7286,[203]8.7456,[204]8.7602,[205]8.7219,[206]8.7232,[207]8.7519,[208]8.7948,[209]8.8008, save_imatrix: stored collected data after 210 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [210]8.8094,[211]8.8229,[212]8.8434,[213]8.8669,[214]8.8710,[215]8.8792,[216]8.8937,[217]8.9263,[218]8.9901,[219]8.9619, save_imatrix: stored collected data after 220 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [220]8.9730,[221]8.9575,[222]8.9700,[223]8.9656,[224]8.9616,[225]8.9832,[226]8.9590,[227]8.9759,[228]8.9843,[229]9.0451, save_imatrix: stored collected data after 230 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat [230]9.1160,[231]9.1884,[232]9.2569,[233]9.3015,[234]9.2740,[235]9.2473,[236]9.2192, save_imatrix: stored collected data after 236 chunks in AutoCoder_S_6.7B-IMat-GGUF/imatrix.dat llama_print_timings: load time = 4638.76 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 = 291757.04 ms / 120832 tokens ( 2.41 ms per token, 414.15 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 = 300310.61 ms / 120833 tokens Final estimate: PPL = 9.2192 +/- 0.10306