build: 3906 (7eee341b) with Apple clang version 15.0.0 (clang-1500.3.9.4) for arm64-apple-darwin23.6.0 llama_model_loader: loaded meta data with 35 key-value pairs and 219 tensors from salamandra-2b-instruct_IQ4_NL.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.type str = model llama_model_loader: - kv 2: general.size_label str = 2.3B llama_model_loader: - kv 3: general.license str = apache-2.0 llama_model_loader: - kv 4: general.tags arr[str,1] = ["text-generation"] llama_model_loader: - kv 5: general.languages arr[str,36] = ["bg", "ca", "code", "cs", "cy", "da"... llama_model_loader: - kv 6: llama.block_count u32 = 24 llama_model_loader: - kv 7: llama.context_length u32 = 8192 llama_model_loader: - kv 8: llama.embedding_length u32 = 2048 llama_model_loader: - kv 9: llama.feed_forward_length u32 = 5440 llama_model_loader: - kv 10: llama.attention.head_count u32 = 16 llama_model_loader: - kv 11: llama.attention.head_count_kv u32 = 16 llama_model_loader: - kv 12: llama.rope.freq_base f32 = 10000.000000 llama_model_loader: - kv 13: llama.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 14: general.file_type u32 = 25 llama_model_loader: - kv 15: llama.vocab_size u32 = 256000 llama_model_loader: - kv 16: llama.rope.dimension_count u32 = 128 llama_model_loader: - kv 17: tokenizer.ggml.add_space_prefix bool = true llama_model_loader: - kv 18: tokenizer.ggml.model str = llama llama_model_loader: - kv 19: tokenizer.ggml.pre str = default llama_model_loader: - kv 20: tokenizer.ggml.tokens arr[str,256000] = ["", "~~", "~~", "", "<|... llama_model_loader: - kv 21: tokenizer.ggml.scores arr[f32,256000] = [-1000.000000, -1000.000000, -1000.00... llama_model_loader: - kv 22: tokenizer.ggml.token_type arr[i32,256000] = [3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, ... llama_model_loader: - kv 23: tokenizer.ggml.bos_token_id u32 = 1 llama_model_loader: - kv 24: tokenizer.ggml.eos_token_id u32 = 2 llama_model_loader: - kv 25: tokenizer.ggml.unknown_token_id u32 = 0 llama_model_loader: - kv 26: tokenizer.ggml.padding_token_id u32 = 0 llama_model_loader: - kv 27: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 28: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 29: tokenizer.chat_template str = {%- if not date_string is defined %}{... llama_model_loader: - kv 30: general.quantization_version u32 = 2 llama_model_loader: - kv 31: quantize.imatrix.file str = imatrix/oscar/imatrix.dat llama_model_loader: - kv 32: quantize.imatrix.dataset str = ./imatrix/oscar/imatrix-dataset.txt llama_model_loader: - kv 33: quantize.imatrix.entries_count i32 = 168 llama_model_loader: - kv 34: quantize.imatrix.chunks_count i32 = 44176 llama_model_loader: - type f32: 49 tensors llama_model_loader: - type iq4_nl: 169 tensors llama_model_loader: - type bf16: 1 tensors llm_load_vocab: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect llm_load_vocab: special tokens cache size = 104 llm_load_vocab: token to piece cache size = 1.8842 MB llm_load_print_meta: format = GGUF V3 (latest) llm_load_print_meta: arch = llama 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 = 2048 llm_load_print_meta: n_layer = 24 llm_load_print_meta: n_head = 16 llm_load_print_meta: n_head_kv = 16 llm_load_print_meta: n_rot = 128 llm_load_print_meta: n_swa = 0 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 = 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-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 = 5440 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 = 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 = ?B llm_load_print_meta: model ftype = IQ4_NL - 4.5 bpw llm_load_print_meta: model params = 2.25 B llm_load_print_meta: model size = 1.88 GiB (7.18 BPW) llm_load_print_meta: general.name = n/a llm_load_print_meta: BOS token = 1 '~~' llm_load_print_meta: EOS token = 2 '~~' llm_load_print_meta: UNK token = 0 '' llm_load_print_meta: PAD token = 0 '' llm_load_print_meta: LF token = 145 '<0x0A>' llm_load_print_meta: EOT token = 5 '<|im_end|>' llm_load_print_meta: EOG token = 2 '' llm_load_print_meta: EOG token = 5 '<|im_end|>' llm_load_print_meta: max token length = 72 llm_load_tensors: ggml ctx size = 0.20 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: Metal buffer size = 1927.95 MiB llm_load_tensors: CPU buffer size = 281.25 MiB .................................... llama_new_context_with_model: n_ctx = 8192 llama_new_context_with_model: n_batch = 512 llama_new_context_with_model: n_ubatch = 128 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 ggml_metal_init: allocating ggml_metal_init: found device: Apple M3 Max ggml_metal_init: picking default device: Apple M3 Max ggml_metal_init: using embedded metal library ggml_metal_init: GPU name: Apple M3 Max ggml_metal_init: GPU family: MTLGPUFamilyApple9 (1009) ggml_metal_init: GPU family: MTLGPUFamilyCommon3 (3003) ggml_metal_init: GPU family: MTLGPUFamilyMetal3 (5001) ggml_metal_init: simdgroup reduction support = true ggml_metal_init: simdgroup matrix mul. support = true ggml_metal_init: hasUnifiedMemory = true ggml_metal_init: recommendedMaxWorkingSetSize = 42949.67 MB llama_kv_cache_init: Metal KV buffer size = 1536.00 MiB llama_new_context_with_model: KV self size = 1536.00 MiB, K (f16): 768.00 MiB, V (f16): 768.00 MiB llama_new_context_with_model: CPU output buffer size = 0.98 MiB llama_new_context_with_model: Metal compute buffer size = 72.00 MiB llama_new_context_with_model: CPU compute buffer size = 125.00 MiB llama_new_context_with_model: graph nodes = 774 llama_new_context_with_model: graph splits = 3 common_init_from_params: warming up the model with an empty run - please wait ... (--no-warmup to disable) system_info: n_threads = 15 (n_threads_batch = 15) / 16 | AVX = 0 | AVX_VNNI = 0 | AVX2 = 0 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | AVX512_BF16 = 0 | FMA = 0 | NEON = 1 | SVE = 0 | ARM_FMA = 1 | F16C = 0 | FP16_VA = 1 | RISCV_VECT = 0 | WASM_SIMD = 0 | BLAS = 1 | SSE3 = 0 | SSSE3 = 0 | VSX = 0 | MATMUL_INT8 = 1 | LLAMAFILE = 1 | perplexity: tokenizing the input .. perplexity: tokenization took 2930.16 ms perplexity: calculating perplexity over 134 chunks, n_ctx=8192, batch_size=512, n_seq=1 perplexity: 10.21 seconds per pass - ETA 22.80 minutes [1]17.8129,[2]18.1041,[3]16.4180,[4]16.1482,[5]15.4605,[6]14.9659,[7]15.8652,[8]15.3564,[9]15.0572,[10]14.3328,[11]14.9764,[12]15.0290,[13]16.1215,[14]16.4023,[15]16.3759,[16]16.9183,[17]17.2192,[18]17.1140,[19]17.1522,[20]17.4642,[21]17.4981,[22]15.4612,[23]15.6374,[24]15.2560,[25]14.7359,[26]14.2838,[27]14.0927,[28]13.9146,[29]13.8618,[30]13.6526,[31]13.8921,[32]13.9902,[33]14.4671,[34]14.7690,[35]15.0739,[36]14.8319,[37]14.8161,[38]14.8899,[39]14.7338,[40]14.7628,[41]14.7376,[42]14.5433,[43]14.4871,[44]14.6541,[45]14.8603,[46]14.7111,[47]14.9554,[48]15.0698,[49]15.3552,[50]15.6439,[51]15.6804,[52]15.8969,[53]16.2215,[54]16.5492,[55]16.6635,[56]16.4929,[57]16.3979,[58]16.1194,[59]16.0097,[60]15.8135,[61]15.8653,[62]16.0048,[63]16.1949,[64]16.2594,[65]16.2889,[66]16.4813,[67]16.4558,[68]16.3421,[69]16.1984,[70]16.0900,[71]16.0856,[72]16.0296,[73]16.0389,[74]15.9815,[75]15.9688,[76]15.9092,[77]15.9666,[78]15.9660,[79]15.9735,[80]16.0090,[81]15.7020,[82]15.6750,[83]15.5430,[84]15.5791,[85]15.6284,[86]15.8285,[87]15.8571,[88]16.0137,[89]16.0674,[90]16.1958,[91]16.2560,[92]16.0914,[93]16.1578,[94]16.1434,[95]16.2849,[96]16.4783,[97]16.5552,[98]16.6532,[99]16.8006,[100]16.8424,[101]16.8692,[102]16.8301,[103]16.8010,[104]16.7855,[105]16.7688,[106]16.6373,[107]16.5060,[108]16.5673,[109]16.5854,[110]16.4956,[111]16.4596,[112]16.3089,[113]16.1649,[114]16.1582,[115]16.1324,[116]16.1411,[117]16.0320,[118]15.8980,[119]15.8904,[120]15.9506,[121]15.9657,[122]15.9907,[123]16.0273,[124]16.0451,[125]16.0401,[126]16.0667,[127]16.0905,[128]16.1714,[129]16.1614,[130]16.1365,[131]16.1945,[132]16.1701,[133]16.1129,[134]15.9602, Final estimate: PPL = 15.9602 +/- 0.06509 llama_perf_context_print: load time = 1276.58 ms llama_perf_context_print: prompt eval time = 1349267.03 ms / 1097728 tokens ( 1.23 ms per token, 813.57 tokens per second) llama_perf_context_print: eval time = 0.00 ms / 1 runs ( 0.00 ms per token, inf tokens per second) llama_perf_context_print: total time = 1388038.33 ms / 1097729 tokens ggml_metal_free: deallocating