main: build = 3006 (eaf6e031) main: built with cc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0 for x86_64-linux-gnu main: seed = 1716806623 llama_model_loader: loaded meta data with 22 key-value pairs and 291 tensors from openchat-3.6-8b-20240522-IMat-GGUF/openchat-3.6-8b-20240522.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 = openchat-3.6-8b-20240522 llama_model_loader: - kv 2: llama.block_count u32 = 32 llama_model_loader: - kv 3: llama.context_length u32 = 8192 llama_model_loader: - kv 4: llama.embedding_length u32 = 4096 llama_model_loader: - kv 5: llama.feed_forward_length u32 = 14336 llama_model_loader: - kv 6: llama.attention.head_count u32 = 32 llama_model_loader: - kv 7: llama.attention.head_count_kv u32 = 8 llama_model_loader: - kv 8: llama.rope.freq_base f32 = 500000.000000 llama_model_loader: - kv 9: llama.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 10: general.file_type u32 = 1 llama_model_loader: - kv 11: llama.vocab_size u32 = 128256 llama_model_loader: - kv 12: llama.rope.dimension_count u32 = 128 llama_model_loader: - kv 13: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 14: tokenizer.ggml.pre str = llama-bpe llama_model_loader: - kv 15: tokenizer.ggml.tokens arr[str,128256] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 16: tokenizer.ggml.token_type arr[i32,128256] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 17: tokenizer.ggml.merges arr[str,280147] = ["Ġ Ġ", "Ġ ĠĠĠ", "ĠĠ ĠĠ", "... llama_model_loader: - kv 18: tokenizer.ggml.bos_token_id u32 = 128000 llama_model_loader: - kv 19: tokenizer.ggml.eos_token_id u32 = 128009 llama_model_loader: - kv 20: tokenizer.chat_template str = {{ bos_token }}{% for message in mess... llama_model_loader: - kv 21: general.quantization_version u32 = 2 llama_model_loader: - type f32: 65 tensors llama_model_loader: - type f16: 226 tensors llm_load_vocab: special tokens definition check successful ( 256/128256 ). 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 = 128256 llm_load_print_meta: n_merges = 280147 llm_load_print_meta: n_ctx_train = 8192 llm_load_print_meta: n_embd = 4096 llm_load_print_meta: n_head = 32 llm_load_print_meta: n_head_kv = 8 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 = 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 = 14336 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 = 500000.0 llm_load_print_meta: freq_scale_train = 1 llm_load_print_meta: n_yarn_orig_ctx = 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: model type = 8B llm_load_print_meta: model ftype = F16 llm_load_print_meta: model params = 8.03 B llm_load_print_meta: model size = 14.96 GiB (16.00 BPW) llm_load_print_meta: general.name = openchat-3.6-8b-20240522 llm_load_print_meta: BOS token = 128000 '<|begin_of_text|>' llm_load_print_meta: EOS token = 128009 '<|eot_id|>' llm_load_print_meta: LF token = 128 'Ä' llm_load_print_meta: EOT token = 128009 '<|eot_id|>' 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 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 = 1002.00 MiB llm_load_tensors: CUDA0 buffer size = 14315.02 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 = 500000.0 llama_new_context_with_model: freq_scale = 1 llama_kv_cache_init: CUDA0 KV buffer size = 64.00 MiB llama_new_context_with_model: KV self size = 64.00 MiB, K (f16): 32.00 MiB, V (f16): 32.00 MiB llama_new_context_with_model: CUDA_Host output buffer size = 0.49 MiB llama_new_context_with_model: CUDA0 compute buffer size = 258.50 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 70.503 ms compute_imatrix: computing over 189 chunks with batch_size 512 compute_imatrix: 0.43 seconds per pass - ETA 1.35 minutes [1]6.0348,[2]4.7804,[3]4.3377,[4]5.4029,[5]5.5439,[6]4.7045,[7]5.0585,[8]5.5263,[9]5.7295, save_imatrix: stored collected data after 10 chunks in openchat-3.6-8b-20240522-IMat-GGUF/imatrix.dat [10]5.7404,[11]6.2072,[12]5.9627,[13]6.4471,[14]6.8690,[15]7.1301,[16]7.5406,[17]7.9763,[18]8.1476,[19]7.7801, save_imatrix: stored collected data after 20 chunks in openchat-3.6-8b-20240522-IMat-GGUF/imatrix.dat [20]7.6815,[21]7.4206,[22]6.9722,[23]6.6753,[24]6.5744,[25]6.7881,[26]6.9148,[27]7.0631,[28]7.0700,[29]6.7843, save_imatrix: stored collected data after 30 chunks in openchat-3.6-8b-20240522-IMat-GGUF/imatrix.dat [30]6.6068,[31]6.5076,[32]6.4810,[33]6.4494,[34]6.4542,[35]6.5629,[36]6.6907,[37]6.8445,[38]6.8942,[39]7.0491, save_imatrix: stored collected data after 40 chunks in openchat-3.6-8b-20240522-IMat-GGUF/imatrix.dat [40]7.2272,[41]7.4645,[42]7.5906,[43]7.7612,[44]7.7605,[45]7.7814,[46]7.8920,[47]8.0201,[48]8.0505,[49]8.1362, save_imatrix: stored collected data after 50 chunks in openchat-3.6-8b-20240522-IMat-GGUF/imatrix.dat [50]8.1781,[51]8.2467,[52]8.2276,[53]8.2640,[54]8.2626,[55]8.2506,[56]8.2139,[57]8.2284,[58]8.3023,[59]8.4184, save_imatrix: stored collected data after 60 chunks in openchat-3.6-8b-20240522-IMat-GGUF/imatrix.dat [60]8.5066,[61]8.4251,[62]8.3594,[63]8.2943,[64]8.2607,[65]8.1936,[66]8.1153,[67]8.0173,[68]8.0054,[69]7.9513, save_imatrix: stored collected data after 70 chunks in openchat-3.6-8b-20240522-IMat-GGUF/imatrix.dat [70]7.9827,[71]8.0265,[72]8.0360,[73]8.0271,[74]8.0691,[75]7.9830,[76]7.9466,[77]7.8510,[78]7.8160,[79]7.7771, save_imatrix: stored collected data after 80 chunks in openchat-3.6-8b-20240522-IMat-GGUF/imatrix.dat [80]7.7414,[81]7.6666,[82]7.5859,[83]7.5116,[84]7.5245,[85]7.5625,[86]7.5638,[87]7.5269,[88]7.5130,[89]7.5259, save_imatrix: stored collected data after 90 chunks in openchat-3.6-8b-20240522-IMat-GGUF/imatrix.dat [90]7.5618,[91]7.5482,[92]7.5582,[93]7.5827,[94]7.6174,[95]7.5908,[96]7.6124,[97]7.6184,[98]7.6096,[99]7.6157, save_imatrix: stored collected data after 100 chunks in openchat-3.6-8b-20240522-IMat-GGUF/imatrix.dat [100]7.6085,[101]7.5939,[102]7.5922,[103]7.6202,[104]7.6382,[105]7.6315,[106]7.6544,[107]7.6763,[108]7.6100,[109]7.6095, save_imatrix: stored collected data after 110 chunks in openchat-3.6-8b-20240522-IMat-GGUF/imatrix.dat [110]7.5896,[111]7.5451,[112]7.5181,[113]7.4785,[114]7.4304,[115]7.3814,[116]7.3309,[117]7.2866,[118]7.2428,[119]7.2931, save_imatrix: stored collected data after 120 chunks in openchat-3.6-8b-20240522-IMat-GGUF/imatrix.dat [120]7.3070,[121]7.3335,[122]7.3880,[123]7.4235,[124]7.4814,[125]7.5484,[126]7.6088,[127]7.6668,[128]7.7422,[129]7.8223, save_imatrix: stored collected data after 130 chunks in openchat-3.6-8b-20240522-IMat-GGUF/imatrix.dat [130]7.7887,[131]7.8083,[132]7.8176,[133]7.8402,[134]7.8135,[135]7.8121,[136]7.8459,[137]7.8508,[138]7.8611,[139]7.8849, save_imatrix: stored collected data after 140 chunks in openchat-3.6-8b-20240522-IMat-GGUF/imatrix.dat [140]7.8981,[141]7.8929,[142]7.9079,[143]7.8729,[144]7.8877,[145]7.9220,[146]7.9362,[147]7.9369,[148]7.9497,[149]7.9625, save_imatrix: stored collected data after 150 chunks in openchat-3.6-8b-20240522-IMat-GGUF/imatrix.dat [150]7.9463,[151]7.9322,[152]7.9401,[153]7.9533,[154]8.0047,[155]7.9772,[156]7.9799,[157]8.0199,[158]8.0687,[159]8.1549, save_imatrix: stored collected data after 160 chunks in openchat-3.6-8b-20240522-IMat-GGUF/imatrix.dat [160]8.2253,[161]8.2376,[162]8.2574,[163]8.2741,[164]8.2780,[165]8.3119,[166]8.3076,[167]8.3057,[168]8.3136,[169]8.3367, save_imatrix: stored collected data after 170 chunks in openchat-3.6-8b-20240522-IMat-GGUF/imatrix.dat [170]8.3299,[171]8.3237,[172]8.3376,[173]8.3447,[174]8.3637,[175]8.3477,[176]8.3407,[177]8.3396,[178]8.3359,[179]8.3239, save_imatrix: stored collected data after 180 chunks in openchat-3.6-8b-20240522-IMat-GGUF/imatrix.dat [180]8.3046,[181]8.3154,[182]8.2919,[183]8.3189,[184]8.3258,[185]8.3672,[186]8.3911,[187]8.4162,[188]8.3767,[189]8.3368, save_imatrix: stored collected data after 189 chunks in openchat-3.6-8b-20240522-IMat-GGUF/imatrix.dat llama_print_timings: load time = 2087.30 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 = 65137.26 ms / 96768 tokens ( 0.67 ms per token, 1485.60 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 = 67989.82 ms / 96769 tokens Final estimate: PPL = 8.3368 +/- 0.10059