llama_model_loader: loaded meta data with 34 key-value pairs and 291 tensors from xLAM-7b-r-IMat-GGUF/xLAM-7b-r.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 = llama llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = xLAM 7b R llama_model_loader: - kv 3: general.finetune str = r llama_model_loader: - kv 4: general.basename str = xLAM llama_model_loader: - kv 5: general.size_label str = 7B llama_model_loader: - kv 6: general.license str = cc-by-nc-4.0 llama_model_loader: - kv 7: general.tags arr[str,6] = ["function-calling", "LLM Agent", "to... llama_model_loader: - kv 8: general.languages arr[str,1] = ["en"] llama_model_loader: - kv 9: general.datasets arr[str,1] = ["Salesforce/xlam-function-calling-60k"] llama_model_loader: - kv 10: llama.block_count u32 = 32 llama_model_loader: - kv 11: llama.context_length u32 = 32768 llama_model_loader: - kv 12: llama.embedding_length u32 = 4096 llama_model_loader: - kv 13: llama.feed_forward_length u32 = 14336 llama_model_loader: - kv 14: llama.attention.head_count u32 = 32 llama_model_loader: - kv 15: llama.attention.head_count_kv u32 = 8 llama_model_loader: - kv 16: llama.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 17: llama.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 18: general.file_type u32 = 7 llama_model_loader: - kv 19: llama.vocab_size u32 = 32000 llama_model_loader: - kv 20: llama.rope.dimension_count u32 = 128 llama_model_loader: - kv 21: tokenizer.ggml.add_space_prefix bool = false llama_model_loader: - kv 22: tokenizer.ggml.model str = llama llama_model_loader: - kv 23: tokenizer.ggml.pre str = default llama_model_loader: - kv 24: tokenizer.ggml.tokens arr[str,32000] = ["", "~~", "~~", "<0x00>", "<... llama_model_loader: - kv 25: tokenizer.ggml.scores arr[f32,32000] = [-1000.000000, -1000.000000, -1000.00... llama_model_loader: - kv 26: tokenizer.ggml.token_type arr[i32,32000] = [3, 3, 3, 6, 6, 6, 6, 6, 6, 6, 6, 6, ... llama_model_loader: - kv 27: tokenizer.ggml.bos_token_id u32 = 1 llama_model_loader: - kv 28: tokenizer.ggml.eos_token_id u32 = 2 llama_model_loader: - kv 29: tokenizer.ggml.unknown_token_id u32 = 0 llama_model_loader: - kv 30: tokenizer.ggml.add_bos_token bool = false llama_model_loader: - kv 31: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 32: tokenizer.chat_template str = {{ bos_token }}{% for message in mess... llama_model_loader: - kv 33: general.quantization_version u32 = 2 llama_model_loader: - type f32: 65 tensors llama_model_loader: - type q8_0: 226 tensors llm_load_vocab: special tokens cache size = 3 llm_load_vocab: token to piece cache size = 0.1637 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 = 32000 llm_load_print_meta: n_merges = 0 llm_load_print_meta: vocab_only = 0 llm_load_print_meta: n_ctx_train = 32768 llm_load_print_meta: n_embd = 4096 llm_load_print_meta: n_layer = 32 llm_load_print_meta: n_head = 32 llm_load_print_meta: n_head_kv = 8 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 = 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 = 1000000.0 llm_load_print_meta: freq_scale_train = 1 llm_load_print_meta: n_ctx_orig_yarn = 32768 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 = 7B llm_load_print_meta: model ftype = Q8_0 llm_load_print_meta: model params = 7.24 B llm_load_print_meta: model size = 7.17 GiB (8.50 BPW) llm_load_print_meta: general.name = xLAM 7b R 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: LF token = 13 '<0x0A>' 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 = 0.27 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 = 132.81 MiB llm_load_tensors: CUDA0 buffer size = 7205.83 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 = 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 = 0.12 MiB llama_new_context_with_model: CUDA0 compute buffer size = 81.00 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 (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 | 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 99.008 ms compute_imatrix: computing over 148 chunks with batch_size 512 compute_imatrix: 0.67 seconds per pass - ETA 1.65 minutes [1]4.0886,[2]2.9452,[3]2.9743,[4]3.1523,[5]3.5458,[6]3.4611,[7]3.1804,[8]3.6540,[9]3.8055, save_imatrix: stored collected data after 10 chunks in xLAM-7b-r-IMat-GGUF/imatrix.dat [10]4.2065,[11]4.3688,[12]4.1191,[13]4.3348,[14]4.5837,[15]4.9394,[16]5.0901,[17]5.3093,[18]5.4405,[19]5.5103, save_imatrix: stored collected data after 20 chunks in xLAM-7b-r-IMat-GGUF/imatrix.dat [20]5.6458,[21]5.5826,[22]5.3903,[23]5.5240,[24]5.5161,[25]5.5700,[26]5.4276,[27]5.6431,[28]5.5657,[29]5.6590, save_imatrix: stored collected data after 30 chunks in xLAM-7b-r-IMat-GGUF/imatrix.dat [30]5.5322,[31]5.6113,[32]5.7234,[33]5.8611,[34]5.8932,[35]5.8244,[36]5.6129,[37]5.4541,[38]5.3355,[39]5.2405, save_imatrix: stored collected data after 40 chunks in xLAM-7b-r-IMat-GGUF/imatrix.dat [40]5.1727,[41]5.1787,[42]5.1454,[43]5.1428,[44]5.0991,[45]5.0783,[46]5.0954,[47]5.1514,[48]5.2317,[49]5.2562, save_imatrix: stored collected data after 50 chunks in xLAM-7b-r-IMat-GGUF/imatrix.dat [50]5.3921,[51]5.4991,[52]5.6570,[53]5.7809,[54]5.8987,[55]5.8522,[56]5.8086,[57]5.8774,[58]5.9485,[59]5.9688, save_imatrix: stored collected data after 60 chunks in xLAM-7b-r-IMat-GGUF/imatrix.dat [60]5.9026,[61]5.8872,[62]5.9106,[63]5.9532,[64]6.0496,[65]6.1234,[66]6.1502,[67]6.1669,[68]6.1966,[69]6.2012, save_imatrix: stored collected data after 70 chunks in xLAM-7b-r-IMat-GGUF/imatrix.dat [70]6.1947,[71]6.1265,[72]6.0751,[73]6.0409,[74]6.0502,[75]6.0795,[76]6.0731,[77]6.0922,[78]6.1034,[79]6.0827, save_imatrix: stored collected data after 80 chunks in xLAM-7b-r-IMat-GGUF/imatrix.dat [80]6.0714,[81]6.0460,[82]6.0652,[83]6.0723,[84]6.0657,[85]6.0826,[86]6.0641,[87]6.0521,[88]6.0342,[89]6.0420, save_imatrix: stored collected data after 90 chunks in xLAM-7b-r-IMat-GGUF/imatrix.dat [90]6.0213,[91]5.9912,[92]5.9652,[93]5.9519,[94]5.9674,[95]5.9917,[96]5.9786,[97]5.9734,[98]5.9635,[99]6.0018, save_imatrix: stored collected data after 100 chunks in xLAM-7b-r-IMat-GGUF/imatrix.dat [100]5.9407,[101]5.9387,[102]5.9168,[103]5.9311,[104]5.9391,[105]5.9253,[106]5.8936,[107]5.8519,[108]5.8104,[109]5.7726, save_imatrix: stored collected data after 110 chunks in xLAM-7b-r-IMat-GGUF/imatrix.dat [110]5.7306,[111]5.6930,[112]5.6564,[113]5.6222,[114]5.5864,[115]5.5595,[116]5.5623,[117]5.5800,[118]5.6310,[119]5.6801, save_imatrix: stored collected data after 120 chunks in xLAM-7b-r-IMat-GGUF/imatrix.dat [120]5.7265,[121]5.7994,[122]5.8640,[123]5.8657,[124]5.8758,[125]5.8380,[126]5.8276,[127]5.8218,[128]5.8240,[129]5.7958, save_imatrix: stored collected data after 130 chunks in xLAM-7b-r-IMat-GGUF/imatrix.dat [130]5.7623,[131]5.7831,[132]5.8170,[133]5.8136,[134]5.8086,[135]5.8195,[136]5.8427,[137]5.8450,[138]5.8518,[139]5.8687, save_imatrix: stored collected data after 140 chunks in xLAM-7b-r-IMat-GGUF/imatrix.dat [140]5.8721,[141]5.8651,[142]5.9060,[143]5.9412,[144]5.9657,[145]6.0045,[146]6.0364,[147]6.0794,[148]6.1121, save_imatrix: stored collected data after 148 chunks in xLAM-7b-r-IMat-GGUF/imatrix.dat llama_print_timings: load time = 2301.03 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 = 86818.00 ms / 75776 tokens ( 1.15 ms per token, 872.81 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 = 88858.24 ms / 75777 tokens Final estimate: PPL = 6.1121 +/- 0.07381