llama_model_loader: loaded meta data with 39 key-value pairs and 508 tensors from datagemma-rig-27b-it-IMat-GGUF/datagemma-rig-27b-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.type str = model llama_model_loader: - kv 2: general.name str = Rig_27B_Transformers_Checkpoint_15000 llama_model_loader: - kv 3: general.finetune str = it llama_model_loader: - kv 4: general.basename str = datagemma-rig llama_model_loader: - kv 5: general.size_label str = 27B llama_model_loader: - kv 6: general.license str = gemma llama_model_loader: - kv 7: general.base_model.count u32 = 1 llama_model_loader: - kv 8: general.base_model.0.name str = Gemma 2 27b It llama_model_loader: - kv 9: general.base_model.0.organization str = Google llama_model_loader: - kv 10: general.base_model.0.repo_url str = https://huggingface.co/google/gemma-2... llama_model_loader: - kv 11: general.tags arr[str,2] = ["conversational", "text-generation"] llama_model_loader: - kv 12: gemma2.context_length u32 = 8192 llama_model_loader: - kv 13: gemma2.embedding_length u32 = 4608 llama_model_loader: - kv 14: gemma2.block_count u32 = 46 llama_model_loader: - kv 15: gemma2.feed_forward_length u32 = 36864 llama_model_loader: - kv 16: gemma2.attention.head_count u32 = 32 llama_model_loader: - kv 17: gemma2.attention.head_count_kv u32 = 16 llama_model_loader: - kv 18: gemma2.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 19: gemma2.attention.key_length u32 = 128 llama_model_loader: - kv 20: gemma2.attention.value_length u32 = 128 llama_model_loader: - kv 21: general.file_type u32 = 7 llama_model_loader: - kv 22: gemma2.attn_logit_softcapping f32 = 50.000000 llama_model_loader: - kv 23: gemma2.final_logit_softcapping f32 = 30.000000 llama_model_loader: - kv 24: gemma2.attention.sliding_window u32 = 4096 llama_model_loader: - kv 25: tokenizer.ggml.model str = llama llama_model_loader: - kv 26: tokenizer.ggml.pre str = default llama_model_loader: - kv 27: tokenizer.ggml.tokens arr[str,256000] = ["", "", "", "", ... llama_model_loader: - kv 28: tokenizer.ggml.scores arr[f32,256000] = [-1000.000000, -1000.000000, -1000.00... llama_model_loader: - kv 29: tokenizer.ggml.token_type arr[i32,256000] = [3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, ... llama_model_loader: - kv 30: tokenizer.ggml.bos_token_id u32 = 2 llama_model_loader: - kv 31: tokenizer.ggml.eos_token_id u32 = 1 llama_model_loader: - kv 32: tokenizer.ggml.unknown_token_id u32 = 3 llama_model_loader: - kv 33: tokenizer.ggml.padding_token_id u32 = 0 llama_model_loader: - kv 34: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 35: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 36: tokenizer.chat_template str = {{ bos_token }}{% if messages[0]['rol... llama_model_loader: - kv 37: tokenizer.ggml.add_space_prefix bool = false llama_model_loader: - kv 38: general.quantization_version u32 = 2 llama_model_loader: - type f32: 185 tensors llama_model_loader: - type q8_0: 323 tensors llm_load_vocab: special tokens cache size = 249 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: vocab_only = 0 llm_load_print_meta: n_ctx_train = 8192 llm_load_print_meta: n_embd = 4608 llm_load_print_meta: n_layer = 46 llm_load_print_meta: n_head = 32 llm_load_print_meta: n_head_kv = 16 llm_load_print_meta: n_rot = 128 llm_load_print_meta: n_swa = 4096 llm_load_print_meta: n_embd_head_k = 128 llm_load_print_meta: n_embd_head_v = 128 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 = 36864 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: ssm_dt_b_c_rms = 0 llm_load_print_meta: model type = 27B llm_load_print_meta: model ftype = Q8_0 llm_load_print_meta: model params = 27.23 B llm_load_print_meta: model size = 26.94 GiB (8.50 BPW) llm_load_print_meta: general.name = Rig_27B_Transformers_Checkpoint_15000 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 = 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 = 0.45 MiB llm_load_tensors: offloading 37 repeating layers to GPU llm_load_tensors: offloaded 37/47 layers to GPU llm_load_tensors: CPU buffer size = 27591.06 MiB llm_load_tensors: CUDA0 buffer size = 21231.35 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 = 36.00 MiB llama_kv_cache_init: CUDA0 KV buffer size = 148.00 MiB llama_new_context_with_model: KV self size = 184.00 MiB, K (f16): 92.00 MiB, V (f16): 92.00 MiB llama_new_context_with_model: CUDA_Host output buffer size = 0.98 MiB llama_new_context_with_model: CUDA0 compute buffer size = 1704.31 MiB llama_new_context_with_model: CUDA_Host compute buffer size = 11.01 MiB llama_new_context_with_model: graph nodes = 1850 llama_new_context_with_model: graph splits = 121 system_info: n_threads = 25 (n_threads_batch = 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 | RISCV_VECT = 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 128.146 ms compute_imatrix: computing over 128 chunks with batch_size 512 compute_imatrix: 2.14 seconds per pass - ETA 4.55 minutes [1]105.3030,[2]48.2459,[3]37.3551,[4]64.6273,[5]58.9392,[6]37.2118,[7]48.0783,[8]55.2094,[9]60.8348, save_imatrix: stored collected data after 10 chunks in datagemma-rig-27b-it-IMat-GGUF/imatrix.dat [10]47.4368,[11]52.2511,[12]62.5349,[13]72.8301,[14]76.6891,[15]88.0492,[16]94.4319,[17]98.4147,[18]102.0178,[19]88.1819, save_imatrix: stored collected data after 20 chunks in datagemma-rig-27b-it-IMat-GGUF/imatrix.dat [20]98.0518,[21]98.3230,[22]98.1572,[23]99.9073,[24]99.7030,[25]107.9212,[26]102.9446,[27]106.7155,[28]111.2190,[29]111.2565, save_imatrix: stored collected data after 30 chunks in datagemma-rig-27b-it-IMat-GGUF/imatrix.dat [30]109.9259,[31]97.0510,[32]90.6971,[33]90.2245,[34]89.5476,[35]90.3216,[36]88.7417,[37]89.0847,[38]91.9367,[39]95.8369, save_imatrix: stored collected data after 40 chunks in datagemma-rig-27b-it-IMat-GGUF/imatrix.dat [40]97.8025,[41]100.7910,[42]105.4919,[43]109.2177,[44]112.0317,[45]113.8144,[46]112.3453,[47]112.6577,[48]117.0320,[49]120.4270, save_imatrix: stored collected data after 50 chunks in datagemma-rig-27b-it-IMat-GGUF/imatrix.dat [50]118.2151,[51]119.8638,[52]120.5760,[53]125.0480,[54]129.1875,[55]130.9818,[56]131.6318,[57]132.9954,[58]133.8396,[59]129.0727, save_imatrix: stored collected data after 60 chunks in datagemma-rig-27b-it-IMat-GGUF/imatrix.dat [60]126.6368,[61]123.9368,[62]125.3806,[63]126.4441,[64]127.2459,[65]127.0590,[66]127.9807,[67]126.9014,[68]125.0932,[69]126.0750, save_imatrix: stored collected data after 70 chunks in datagemma-rig-27b-it-IMat-GGUF/imatrix.dat [70]125.7800,[71]125.5559,[72]125.7992,[73]126.4120,[74]125.5738,[75]124.9096,[76]124.8906,[77]125.2136,[78]125.9563,[79]124.9185, save_imatrix: stored collected data after 80 chunks in datagemma-rig-27b-it-IMat-GGUF/imatrix.dat [80]127.0135,[81]128.7838,[82]128.5292,[83]129.6607,[84]131.4402,[85]126.4915,[86]126.0314,[87]124.2917,[88]124.8511,[89]125.4370, save_imatrix: stored collected data after 90 chunks in datagemma-rig-27b-it-IMat-GGUF/imatrix.dat [90]125.3858,[91]123.5666,[92]121.7223,[93]119.4879,[94]117.2487,[95]115.7179,[96]114.0167,[97]112.5808,[98]110.9505,[99]111.4794, save_imatrix: stored collected data after 100 chunks in datagemma-rig-27b-it-IMat-GGUF/imatrix.dat [100]112.0440,[101]114.0937,[102]115.4479,[103]117.9564,[104]120.7776,[105]123.3646,[106]123.9526,[107]124.5521,[108]124.0001,[109]123.4625, save_imatrix: stored collected data after 110 chunks in datagemma-rig-27b-it-IMat-GGUF/imatrix.dat [110]122.9281,[111]121.4704,[112]119.4281,[113]121.1361,[114]122.5180,[115]122.2572,[116]122.8735,[117]124.5789,[118]124.8416,[119]125.0191, save_imatrix: stored collected data after 120 chunks in datagemma-rig-27b-it-IMat-GGUF/imatrix.dat [120]125.5826,[121]126.2783,[122]125.2195,[123]125.7870,[124]126.1228,[125]126.8911,[126]127.9069,[127]129.0043,[128]128.6395, save_imatrix: stored collected data after 128 chunks in datagemma-rig-27b-it-IMat-GGUF/imatrix.dat llama_perf_print: load time = 5108.73 ms llama_perf_print: prompt eval time = 224316.01 ms / 65536 tokens ( 3.42 ms per token, 292.16 tokens per second) llama_perf_print: eval time = 0.00 ms / 1 runs ( 0.00 ms per token, inf tokens per second) llama_perf_print: total time = 228887.72 ms / 65537 tokens Final estimate: PPL = 128.6395 +/- 4.82771