main: build = 3086 (554c247c) main: built with cc (Ubuntu 13.2.0-23ubuntu4) 13.2.0 for x86_64-linux-gnu main: seed = 1717697873 llama_model_loader: loaded meta data with 21 key-value pairs and 290 tensors from Qwen2-0.5B-IMat-GGUF/Qwen2-0.5B.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 = qwen2 llama_model_loader: - kv 1: general.name str = Qwen2-0.5B llama_model_loader: - kv 2: qwen2.block_count u32 = 24 llama_model_loader: - kv 3: qwen2.context_length u32 = 131072 llama_model_loader: - kv 4: qwen2.embedding_length u32 = 896 llama_model_loader: - kv 5: qwen2.feed_forward_length u32 = 4864 llama_model_loader: - kv 6: qwen2.attention.head_count u32 = 14 llama_model_loader: - kv 7: qwen2.attention.head_count_kv u32 = 2 llama_model_loader: - kv 8: qwen2.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 9: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 10: general.file_type u32 = 0 llama_model_loader: - kv 11: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 12: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 13: tokenizer.ggml.tokens arr[str,151936] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 14: tokenizer.ggml.token_type arr[i32,151936] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 15: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 16: tokenizer.ggml.eos_token_id u32 = 151643 llama_model_loader: - kv 17: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 18: tokenizer.ggml.bos_token_id u32 = 151643 llama_model_loader: - kv 19: tokenizer.chat_template str = {% for message in messages %}{% if lo... llama_model_loader: - kv 20: general.quantization_version u32 = 2 llama_model_loader: - type f32: 290 tensors llm_load_vocab: special tokens cache size = 293 llm_load_vocab: token to piece cache size = 0.9338 MB llm_load_print_meta: format = GGUF V3 (latest) llm_load_print_meta: arch = qwen2 llm_load_print_meta: vocab type = BPE llm_load_print_meta: n_vocab = 151936 llm_load_print_meta: n_merges = 151387 llm_load_print_meta: n_ctx_train = 131072 llm_load_print_meta: n_embd = 896 llm_load_print_meta: n_head = 14 llm_load_print_meta: n_head_kv = 2 llm_load_print_meta: n_layer = 24 llm_load_print_meta: n_rot = 64 llm_load_print_meta: n_embd_head_k = 64 llm_load_print_meta: n_embd_head_v = 64 llm_load_print_meta: n_gqa = 7 llm_load_print_meta: n_embd_k_gqa = 128 llm_load_print_meta: n_embd_v_gqa = 128 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 = 4864 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 = 2 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_yarn_orig_ctx = 131072 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 = 1B llm_load_print_meta: model ftype = all F32 llm_load_print_meta: model params = 494.03 M llm_load_print_meta: model size = 1.84 GiB (32.00 BPW) llm_load_print_meta: general.name = Qwen2-0.5B llm_load_print_meta: BOS token = 151643 '<|endoftext|>' llm_load_print_meta: EOS token = 151643 '<|endoftext|>' llm_load_print_meta: PAD token = 151643 '<|endoftext|>' llm_load_print_meta: LF token = 148848 'ÄĬ' llm_load_print_meta: EOT token = 151645 '<|im_end|>' 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 = 0.28 MiB llm_load_tensors: offloading 24 repeating layers to GPU llm_load_tensors: offloading non-repeating layers to GPU llm_load_tensors: offloaded 25/25 layers to GPU llm_load_tensors: CPU buffer size = 519.31 MiB llm_load_tensors: CUDA0 buffer size = 1884.59 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 = 6.00 MiB llama_new_context_with_model: KV self size = 6.00 MiB, K (f16): 3.00 MiB, V (f16): 3.00 MiB llama_new_context_with_model: CUDA_Host output buffer size = 0.58 MiB llama_new_context_with_model: CUDA0 compute buffer size = 298.50 MiB llama_new_context_with_model: CUDA_Host compute buffer size = 2.76 MiB llama_new_context_with_model: graph nodes = 846 llama_new_context_with_model: graph splits = 2 system_info: n_threads = 25 / 32 | AVX = 1 | AVX_VNNI = 0 | AVX2 = 1 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | AVX512_BF16 = 0 | 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 140.904 ms compute_imatrix: computing over 128 chunks with batch_size 512 compute_imatrix: 0.28 seconds per pass - ETA 0.58 minutes [1]10.1721,[2]8.3176,[3]7.5172,[4]9.2414,[5]8.9969,[6]8.1977,[7]8.8937,[8]9.2658,[9]9.9803, save_imatrix: stored collected data after 10 chunks in Qwen2-0.5B-IMat-GGUF/imatrix.dat [10]9.2031,[11]8.8157,[12]9.5016,[13]10.4404,[14]10.6447,[15]11.3873,[16]11.8833,[17]11.9791,[18]12.5532,[19]12.0327, save_imatrix: stored collected data after 20 chunks in Qwen2-0.5B-IMat-GGUF/imatrix.dat [20]12.0846,[21]12.1975,[22]12.2027,[23]12.0807,[24]12.4923,[25]12.7013,[26]12.6513,[27]13.0627,[28]13.4163,[29]13.8037, save_imatrix: stored collected data after 30 chunks in Qwen2-0.5B-IMat-GGUF/imatrix.dat [30]13.7061,[31]13.1443,[32]12.5288,[33]12.1655,[34]11.9420,[35]11.6796,[36]11.7890,[37]12.0546,[38]12.2444,[39]12.2581, save_imatrix: stored collected data after 40 chunks in Qwen2-0.5B-IMat-GGUF/imatrix.dat [40]12.4652,[41]12.4853,[42]13.0221,[43]13.5130,[44]13.9855,[45]14.3595,[46]14.5631,[47]14.3458,[48]14.3392,[49]14.4412, save_imatrix: stored collected data after 50 chunks in Qwen2-0.5B-IMat-GGUF/imatrix.dat [50]14.4927,[51]14.3272,[52]14.3899,[53]14.6646,[54]14.8202,[55]15.0148,[56]15.0869,[57]15.0882,[58]15.1209,[59]15.1033, save_imatrix: stored collected data after 60 chunks in Qwen2-0.5B-IMat-GGUF/imatrix.dat [60]15.1090,[61]14.9807,[62]14.9169,[63]14.9857,[64]15.0805,[65]14.9817,[66]14.9546,[67]14.9305,[68]14.8649,[69]14.8360, save_imatrix: stored collected data after 70 chunks in Qwen2-0.5B-IMat-GGUF/imatrix.dat [70]14.8322,[71]14.7522,[72]14.6928,[73]14.7024,[74]14.5408,[75]14.4171,[76]14.3052,[77]14.2548,[78]14.2444,[79]14.1985, save_imatrix: stored collected data after 80 chunks in Qwen2-0.5B-IMat-GGUF/imatrix.dat [80]14.1018,[81]14.1258,[82]14.0997,[83]13.9914,[84]14.0604,[85]14.0948,[86]13.9901,[87]13.9129,[88]13.8585,[89]13.8845, save_imatrix: stored collected data after 90 chunks in Qwen2-0.5B-IMat-GGUF/imatrix.dat [90]13.8996,[91]13.8839,[92]13.7192,[93]13.5568,[94]13.3917,[95]13.2385,[96]13.0970,[97]12.9408,[98]12.7971,[99]12.7518, save_imatrix: stored collected data after 100 chunks in Qwen2-0.5B-IMat-GGUF/imatrix.dat [100]12.7727,[101]12.8037,[102]12.9817,[103]13.1216,[104]13.2440,[105]13.4238,[106]13.5376,[107]13.5898,[108]13.5245,[109]13.5001, save_imatrix: stored collected data after 110 chunks in Qwen2-0.5B-IMat-GGUF/imatrix.dat [110]13.5125,[111]13.4647,[112]13.4106,[113]13.4351,[114]13.4962,[115]13.4848,[116]13.4894,[117]13.4998,[118]13.5375,[119]13.5129, save_imatrix: stored collected data after 120 chunks in Qwen2-0.5B-IMat-GGUF/imatrix.dat [120]13.4830,[121]13.4752,[122]13.3882,[123]13.4591,[124]13.5547,[125]13.6481,[126]13.7664,[127]13.8600,[128]13.9527, save_imatrix: stored collected data after 128 chunks in Qwen2-0.5B-IMat-GGUF/imatrix.dat llama_print_timings: load time = 913.95 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 = 16920.17 ms / 65536 tokens ( 0.26 ms per token, 3873.25 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 = 18546.35 ms / 65537 tokens Final estimate: PPL = 13.9527 +/- 0.21597