llama_model_loader: loaded meta data with 28 key-value pairs and 464 tensors from gemma-2-9b-it-IMat-GGUF/gemma-2-9b-it.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 = gemma2 llama_model_loader: - kv 1: general.name str = gemma-2-9b-it llama_model_loader: - kv 2: gemma2.context_length u32 = 8192 llama_model_loader: - kv 3: gemma2.embedding_length u32 = 3584 llama_model_loader: - kv 4: gemma2.block_count u32 = 42 llama_model_loader: - kv 5: gemma2.feed_forward_length u32 = 14336 llama_model_loader: - kv 6: gemma2.attention.head_count u32 = 16 llama_model_loader: - kv 7: gemma2.attention.head_count_kv u32 = 8 llama_model_loader: - kv 8: gemma2.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 9: gemma2.attention.key_length u32 = 256 llama_model_loader: - kv 10: gemma2.attention.value_length u32 = 256 llama_model_loader: - kv 11: general.file_type u32 = 7 llama_model_loader: - kv 12: gemma2.attn_logit_softcapping f32 = 50.000000 llama_model_loader: - kv 13: gemma2.final_logit_softcapping f32 = 30.000000 llama_model_loader: - kv 14: tokenizer.ggml.model str = llama llama_model_loader: - kv 15: tokenizer.ggml.pre str = default llama_model_loader: - kv 16: tokenizer.ggml.tokens arr[str,256000] = ["", "", "", "", ... llama_model_loader: - kv 17: tokenizer.ggml.scores arr[f32,256000] = [-1000.000000, -1000.000000, -1000.00... llama_model_loader: - kv 18: tokenizer.ggml.token_type arr[i32,256000] = [3, 3, 3, 3, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 2 llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 1 llama_model_loader: - kv 21: tokenizer.ggml.unknown_token_id u32 = 3 llama_model_loader: - kv 22: tokenizer.ggml.padding_token_id u32 = 0 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 = {{ bos_token }}{% if messages[0]['rol... llama_model_loader: - kv 26: tokenizer.ggml.add_space_prefix bool = false llama_model_loader: - kv 27: general.quantization_version u32 = 2 llama_model_loader: - type f32: 169 tensors llama_model_loader: - type q8_0: 295 tensors llm_load_vocab: special tokens cache size = 261 llm_load_vocab: token to piece cache size = 1.6014 MB llm_load_print_meta: format = GGUF V3 (latest) llm_load_print_meta: arch = gemma2 llm_load_print_meta: vocab type = SPM llm_load_print_meta: n_vocab = 256000 llm_load_print_meta: n_merges = 0 llm_load_print_meta: n_ctx_train = 8192 llm_load_print_meta: n_embd = 3584 llm_load_print_meta: n_head = 16 llm_load_print_meta: n_head_kv = 8 llm_load_print_meta: n_layer = 42 llm_load_print_meta: n_rot = 224 llm_load_print_meta: n_embd_head_k = 256 llm_load_print_meta: n_embd_head_v = 256 llm_load_print_meta: n_gqa = 2 llm_load_print_meta: n_embd_k_gqa = 2048 llm_load_print_meta: n_embd_v_gqa = 2048 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 = 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 = 2 llm_load_print_meta: rope scaling = linear llm_load_print_meta: freq_base_train = 10000.0 llm_load_print_meta: freq_scale_train = 1 llm_load_print_meta: n_ctx_orig_yarn = 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 = 9B llm_load_print_meta: model ftype = Q8_0 llm_load_print_meta: model params = 9.24 B llm_load_print_meta: model size = 9.15 GiB (8.50 BPW) llm_load_print_meta: general.name = gemma-2-9b-it llm_load_print_meta: BOS token = 2 '' llm_load_print_meta: EOS token = 1 '' llm_load_print_meta: UNK token = 3 '' llm_load_print_meta: PAD token = 0 '' llm_load_print_meta: LF token = 227 '<0x0A>' llm_load_print_meta: EOT token = 107 '' llm_load_print_meta: max token length = 93 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 = 0.41 MiB llm_load_tensors: offloading 11 repeating layers to GPU llm_load_tensors: offloaded 11/43 layers to GPU llm_load_tensors: CPU buffer size = 9366.12 MiB llm_load_tensors: CUDA0 buffer size = 2209.54 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 = 10000.0 llama_new_context_with_model: freq_scale = 1 llama_kv_cache_init: CUDA_Host KV buffer size = 124.00 MiB llama_kv_cache_init: CUDA0 KV buffer size = 44.00 MiB llama_new_context_with_model: KV self size = 168.00 MiB, K (f16): 84.00 MiB, V (f16): 84.00 MiB llama_new_context_with_model: CUDA_Host output buffer size = 0.98 MiB llama_new_context_with_model: CUDA0 compute buffer size = 1436.69 MiB llama_new_context_with_model: CUDA_Host compute buffer size = 9.01 MiB llama_new_context_with_model: graph nodes = 1690 llama_new_context_with_model: graph splits = 407 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 91.303 ms compute_imatrix: computing over 128 chunks with batch_size 512 compute_imatrix: 1.34 seconds per pass - ETA 2.85 minutes [1]14.7486,[2]7.3963,[3]6.5677,[4]7.8490,[5]8.6193,[6]9.1006,[7]10.0154,[8]10.8556,[9]11.1957, save_imatrix: stored collected data after 10 chunks in gemma-2-9b-it-IMat-GGUF/imatrix.dat [10]9.8358,[11]9.5880,[12]10.6132,[13]11.2166,[14]11.3352,[15]12.0575,[16]12.1605,[17]12.2130,[18]12.6384,[19]12.5495, save_imatrix: stored collected data after 20 chunks in gemma-2-9b-it-IMat-GGUF/imatrix.dat [20]12.7233,[21]13.7801,[22]13.8033,[23]13.5919,[24]13.8287,[25]13.6927,[26]13.4495,[27]13.6978,[28]13.9355,[29]13.9524, save_imatrix: stored collected data after 30 chunks in gemma-2-9b-it-IMat-GGUF/imatrix.dat [30]14.1488,[31]13.1149,[32]12.5387,[33]12.1100,[34]11.7329,[35]11.4517,[36]11.6110,[37]11.9132,[38]12.0730,[39]12.2426, save_imatrix: stored collected data after 40 chunks in gemma-2-9b-it-IMat-GGUF/imatrix.dat [40]12.3500,[41]12.3930,[42]12.9499,[43]13.3055,[44]13.6998,[45]13.9306,[46]13.6677,[47]13.4581,[48]13.6644,[49]13.8789, save_imatrix: stored collected data after 50 chunks in gemma-2-9b-it-IMat-GGUF/imatrix.dat [50]13.6841,[51]13.5410,[52]13.6019,[53]13.8284,[54]14.1067,[55]14.3488,[56]14.4667,[57]14.4566,[58]14.4479,[59]14.2384, save_imatrix: stored collected data after 60 chunks in gemma-2-9b-it-IMat-GGUF/imatrix.dat [60]14.0794,[61]13.9048,[62]13.7419,[63]13.8516,[64]13.9902,[65]13.8558,[66]13.8803,[67]13.8491,[68]13.7963,[69]13.7287, save_imatrix: stored collected data after 70 chunks in gemma-2-9b-it-IMat-GGUF/imatrix.dat [70]13.6754,[71]13.6579,[72]13.6290,[73]13.6891,[74]13.6254,[75]13.5058,[76]13.4907,[77]13.5002,[78]13.4537,[79]13.3696, save_imatrix: stored collected data after 80 chunks in gemma-2-9b-it-IMat-GGUF/imatrix.dat [80]13.4245,[81]13.4999,[82]13.5207,[83]13.6156,[84]13.6529,[85]13.4393,[86]13.3795,[87]13.2302,[88]13.2605,[89]13.2322, save_imatrix: stored collected data after 90 chunks in gemma-2-9b-it-IMat-GGUF/imatrix.dat [90]13.2980,[91]13.2455,[92]13.1607,[93]13.0864,[94]12.9782,[95]12.9127,[96]12.8267,[97]12.7595,[98]12.6714,[99]12.7111, save_imatrix: stored collected data after 100 chunks in gemma-2-9b-it-IMat-GGUF/imatrix.dat [100]12.7194,[101]12.8597,[102]12.9495,[103]13.0106,[104]13.1816,[105]13.3131,[106]13.3254,[107]13.3336,[108]13.2762,[109]13.3038, save_imatrix: stored collected data after 110 chunks in gemma-2-9b-it-IMat-GGUF/imatrix.dat [110]13.1821,[111]13.0499,[112]12.8931,[113]12.9668,[114]13.0116,[115]12.9995,[116]12.9720,[117]13.0261,[118]13.0587,[119]13.0765, save_imatrix: stored collected data after 120 chunks in gemma-2-9b-it-IMat-GGUF/imatrix.dat [120]13.0687,[121]13.0658,[122]13.0268,[123]13.0554,[124]13.1495,[125]13.2406,[126]13.3502,[127]13.3960,[128]13.4523, save_imatrix: stored collected data after 128 chunks in gemma-2-9b-it-IMat-GGUF/imatrix.dat llama_print_timings: load time = 1979.52 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 = 150211.73 ms / 65536 tokens ( 2.29 ms per token, 436.29 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 = 152551.51 ms / 65537 tokens Final estimate: PPL = 13.4523 +/- 0.26031