main: build = 3008 (1d8fca72) main: built with cc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0 for x86_64-linux-gnu main: seed = 1716818940 llama_model_loader: loaded meta data with 22 key-value pairs and 291 tensors from internlm2-math-plus-7b-IMat-GGUF/internlm2-math-plus-7b.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 = internlm2 llama_model_loader: - kv 1: general.name str = InternLM2 llama_model_loader: - kv 2: internlm2.context_length u32 = 8192 llama_model_loader: - kv 3: internlm2.block_count u32 = 32 llama_model_loader: - kv 4: internlm2.embedding_length u32 = 4096 llama_model_loader: - kv 5: internlm2.feed_forward_length u32 = 14336 llama_model_loader: - kv 6: internlm2.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 7: internlm2.attention.head_count u32 = 32 llama_model_loader: - kv 8: internlm2.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 9: internlm2.attention.head_count_kv u32 = 8 llama_model_loader: - kv 10: general.file_type u32 = 0 llama_model_loader: - kv 11: tokenizer.ggml.model str = llama llama_model_loader: - kv 12: tokenizer.ggml.pre str = default llama_model_loader: - kv 13: tokenizer.ggml.tokens arr[str,92544] = ["", "~~", "~~", "<0x00>", "<... llama_model_loader: - kv 14: tokenizer.ggml.scores arr[f32,92544] = [0.000000, 0.000000, 0.000000, 0.0000... llama_model_loader: - kv 15: tokenizer.ggml.token_type arr[i32,92544] = [2, 3, 3, 6, 6, 6, 6, 6, 6, 6, 6, 6, ... llama_model_loader: - kv 16: tokenizer.ggml.add_space_prefix bool = false llama_model_loader: - kv 17: tokenizer.ggml.bos_token_id u32 = 1 llama_model_loader: - kv 18: tokenizer.ggml.eos_token_id u32 = 2 llama_model_loader: - kv 19: tokenizer.ggml.padding_token_id u32 = 2 llama_model_loader: - kv 20: tokenizer.chat_template str = {{ bos_token }}{% for message in mess... llama_model_loader: - kv 21: general.quantization_version u32 = 2 llama_model_loader: - type f32: 291 tensors llm_load_vocab: mismatch in special tokens definition ( 405/92544 vs 259/92544 ). llm_load_print_meta: format = GGUF V3 (latest) llm_load_print_meta: arch = internlm2 llm_load_print_meta: vocab type = SPM llm_load_print_meta: n_vocab = 92544 llm_load_print_meta: n_merges = 0 llm_load_print_meta: n_ctx_train = 8192 llm_load_print_meta: n_embd = 4096 llm_load_print_meta: n_head = 32 llm_load_print_meta: n_head_kv = 8 llm_load_print_meta: n_layer = 32 llm_load_print_meta: n_rot = 128 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_yarn_orig_ctx = 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 = 7B llm_load_print_meta: model ftype = all F32 llm_load_print_meta: model params = 7.74 B llm_load_print_meta: model size = 28.83 GiB (32.00 BPW) llm_load_print_meta: general.name = InternLM2 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 = 2 '' llm_load_print_meta: LF token = 13 '<0x0A>' 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 = 0.30 MiB llm_load_tensors: offloading 24 repeating layers to GPU llm_load_tensors: offloaded 24/33 layers to GPU llm_load_tensors: CPU buffer size = 29517.02 MiB llm_load_tensors: CUDA0 buffer size = 19968.75 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: CUDA_Host KV buffer size = 16.00 MiB llama_kv_cache_init: CUDA0 KV buffer size = 48.00 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 = 0.35 MiB llama_new_context_with_model: CUDA0 compute buffer size = 1634.75 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 = 92 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 145.596 ms compute_imatrix: computing over 209 chunks with batch_size 512 compute_imatrix: 0.96 seconds per pass - ETA 3.32 minutes [1]8.6168,[2]6.0773,[3]5.4163,[4]6.1982,[5]6.3812,[6]5.6271,[7]6.3406,[8]6.3271,[9]7.0202, save_imatrix: stored collected data after 10 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat [10]7.2999,[11]7.9082,[12]8.0650,[13]9.5172,[14]9.9489,[15]10.7399,[16]11.0800,[17]11.6268,[18]11.0305,[19]11.5148, save_imatrix: stored collected data after 20 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat [20]11.4900,[21]10.8624,[22]10.9906,[23]10.2198,[24]9.8705,[25]9.3718,[26]9.2935,[27]9.7640,[28]9.7553,[29]10.1653, save_imatrix: stored collected data after 30 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat [30]10.4245,[31]10.3090,[32]9.7674,[33]9.4135,[34]9.3314,[35]9.2103,[36]9.0950,[37]9.3619,[38]9.6158,[39]9.8277, save_imatrix: stored collected data after 40 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat [40]10.1027,[41]10.3507,[42]10.6739,[43]10.9750,[44]11.2557,[45]11.3919,[46]11.4866,[47]11.4222,[48]11.3479,[49]11.5629, save_imatrix: stored collected data after 50 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat [50]11.7936,[51]11.8713,[52]12.1181,[53]12.1751,[54]12.1674,[55]12.0785,[56]12.0195,[57]11.9204,[58]11.9528,[59]11.8437, save_imatrix: stored collected data after 60 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat [60]11.8499,[61]12.0561,[62]12.2897,[63]12.6332,[64]12.6103,[65]12.4497,[66]12.1780,[67]11.9540,[68]11.7427,[69]11.5259, save_imatrix: stored collected data after 70 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat [70]11.4201,[71]11.1605,[72]10.9335,[73]10.6764,[74]10.8084,[75]10.9271,[76]10.9531,[77]10.9341,[78]11.0427,[79]11.0051, save_imatrix: stored collected data after 80 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat [80]11.0146,[81]10.9573,[82]10.9558,[83]10.9465,[84]10.9364,[85]10.9364,[86]10.9115,[87]10.9393,[88]10.9545,[89]11.0222, save_imatrix: stored collected data after 90 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat [90]11.0829,[91]11.1312,[92]11.0959,[93]11.0311,[94]10.9948,[95]10.9958,[96]10.9463,[97]10.9640,[98]10.9649,[99]10.9665, save_imatrix: stored collected data after 100 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat [100]10.9076,[101]10.9234,[102]10.8731,[103]10.7965,[104]10.7553,[105]10.7489,[106]10.6998,[107]10.6522,[108]10.6281,[109]10.6360, save_imatrix: stored collected data after 110 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat [110]10.6586,[111]10.6372,[112]10.6652,[113]10.6926,[114]10.6349,[115]10.5984,[116]10.6427,[117]10.6177,[118]10.6488,[119]10.5550, save_imatrix: stored collected data after 120 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat [120]10.4760,[121]10.3803,[122]10.2836,[123]10.1938,[124]10.1074,[125]10.0241,[126]9.9999,[127]9.9689,[128]9.9235,[129]9.8538, save_imatrix: stored collected data after 130 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat [130]9.8316,[131]9.7957,[132]9.7496,[133]9.7198,[134]9.6710,[135]9.6351,[136]9.6114,[137]9.5686,[138]9.5217,[139]9.5032, save_imatrix: stored collected data after 140 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat [140]9.4517,[141]9.4110,[142]9.5089,[143]9.6658,[144]9.8620,[145]10.0062,[146]10.0122,[147]10.0509,[148]10.0944,[149]10.1520, save_imatrix: stored collected data after 150 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat [150]10.1926,[151]10.2064,[152]10.2056,[153]10.2728,[154]10.3324,[155]10.4128,[156]10.4584,[157]10.5164,[158]10.5840,[159]10.6087, save_imatrix: stored collected data after 160 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat [160]10.6699,[161]10.7003,[162]10.7507,[163]10.7984,[164]10.8557,[165]10.8774,[166]10.9137,[167]10.9503,[168]10.9603,[169]10.9802, save_imatrix: stored collected data after 170 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat [170]11.0021,[171]11.0323,[172]11.0788,[173]11.1006,[174]11.0747,[175]11.1685,[176]11.2557,[177]11.3494,[178]11.4653,[179]11.5544, save_imatrix: stored collected data after 180 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat [180]11.5983,[181]11.5868,[182]11.6085,[183]11.6615,[184]11.7197,[185]11.7194,[186]11.7281,[187]11.7453,[188]11.7771,[189]11.7798, save_imatrix: stored collected data after 190 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat [190]11.7851,[191]11.8087,[192]11.8173,[193]11.8368,[194]11.8177,[195]11.8339,[196]11.8145,[197]11.8451,[198]11.8571,[199]11.9365, save_imatrix: stored collected data after 200 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat [200]11.8686,[201]11.9083,[202]11.8877,[203]11.9848,[204]12.0625,[205]12.1544,[206]12.2099,[207]12.2741,[208]12.2187,[209]12.1662, save_imatrix: stored collected data after 209 chunks in internlm2-math-plus-7b-IMat-GGUF/imatrix.dat llama_print_timings: load time = 2815.28 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 = 184148.18 ms / 107008 tokens ( 1.72 ms per token, 581.10 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 = 187280.86 ms / 107009 tokens Final estimate: PPL = 12.1662 +/- 0.16829