File size: 10,369 Bytes
51a05d1 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 |
build: 3785 (64c6af31) with cc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0 for x86_64-linux-gnu
llama_model_loader: loaded meta data with 35 key-value pairs and 434 tensors from Qwen2.5-3B-Instruct-IMat-GGUF/Qwen2.5-3B-Instruct.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 = qwen2
llama_model_loader: - kv 1: general.type str = model
llama_model_loader: - kv 2: general.name str = Qwen2.5 3B Instruct
llama_model_loader: - kv 3: general.finetune str = Instruct
llama_model_loader: - kv 4: general.basename str = Qwen2.5
llama_model_loader: - kv 5: general.size_label str = 3B
llama_model_loader: - kv 6: general.license str = other
llama_model_loader: - kv 7: general.license.name str = qwen-research
llama_model_loader: - kv 8: general.license.link str = https://huggingface.co/Qwen/Qwen2.5-3...
llama_model_loader: - kv 9: general.base_model.count u32 = 1
llama_model_loader: - kv 10: general.base_model.0.name str = Qwen2.5 3B
llama_model_loader: - kv 11: general.base_model.0.organization str = Qwen
llama_model_loader: - kv 12: general.base_model.0.repo_url str = https://huggingface.co/Qwen/Qwen2.5-3B
llama_model_loader: - kv 13: general.tags arr[str,2] = ["chat", "text-generation"]
llama_model_loader: - kv 14: general.languages arr[str,1] = ["en"]
llama_model_loader: - kv 15: qwen2.block_count u32 = 36
llama_model_loader: - kv 16: qwen2.context_length u32 = 32768
llama_model_loader: - kv 17: qwen2.embedding_length u32 = 2048
llama_model_loader: - kv 18: qwen2.feed_forward_length u32 = 11008
llama_model_loader: - kv 19: qwen2.attention.head_count u32 = 16
llama_model_loader: - kv 20: qwen2.attention.head_count_kv u32 = 2
llama_model_loader: - kv 21: qwen2.rope.freq_base f32 = 1000000.000000
llama_model_loader: - kv 22: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000001
llama_model_loader: - kv 23: general.file_type u32 = 7
llama_model_loader: - kv 24: tokenizer.ggml.model str = gpt2
llama_model_loader: - kv 25: tokenizer.ggml.pre str = qwen2
llama_model_loader: - kv 26: tokenizer.ggml.tokens arr[str,151936] = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv 27: tokenizer.ggml.token_type arr[i32,151936] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv 28: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv 29: tokenizer.ggml.eos_token_id u32 = 151645
llama_model_loader: - kv 30: tokenizer.ggml.padding_token_id u32 = 151643
llama_model_loader: - kv 31: tokenizer.ggml.bos_token_id u32 = 151643
llama_model_loader: - kv 32: tokenizer.ggml.add_bos_token bool = false
llama_model_loader: - kv 33: tokenizer.chat_template str = {%- if tools %}\n {{- '<|im_start|>...
llama_model_loader: - kv 34: general.quantization_version u32 = 2
llama_model_loader: - type f32: 181 tensors
llama_model_loader: - type q8_0: 253 tensors
llm_load_vocab: special tokens cache size = 22
llm_load_vocab: token to piece cache size = 0.9310 MB
llm_load_print_meta: format = GGUF V3 (latest)
llm_load_print_meta: arch = qwen2
llm_load_print_meta: vocab type = BPE
llm_load_print_meta: n_vocab = 151936
llm_load_print_meta: n_merges = 151387
llm_load_print_meta: vocab_only = 0
llm_load_print_meta: n_ctx_train = 32768
llm_load_print_meta: n_embd = 2048
llm_load_print_meta: n_layer = 36
llm_load_print_meta: n_head = 16
llm_load_print_meta: n_head_kv = 2
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 = 8
llm_load_print_meta: n_embd_k_gqa = 256
llm_load_print_meta: n_embd_v_gqa = 256
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 = 11008
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 = 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 = ?B
llm_load_print_meta: model ftype = Q8_0
llm_load_print_meta: model params = 3.09 B
llm_load_print_meta: model size = 3.05 GiB (8.50 BPW)
llm_load_print_meta: general.name = Qwen2.5 3B Instruct
llm_load_print_meta: BOS token = 151643 '<|endoftext|>'
llm_load_print_meta: EOS token = 151645 '<|im_end|>'
llm_load_print_meta: PAD token = 151643 '<|endoftext|>'
llm_load_print_meta: LF token = 148848 'ÄĬ'
llm_load_print_meta: EOT token = 151645 '<|im_end|>'
llm_load_print_meta: max token length = 256
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 = 0.38 MiB
llm_load_tensors: offloading 36 repeating layers to GPU
llm_load_tensors: offloading non-repeating layers to GPU
llm_load_tensors: offloaded 37/37 layers to GPU
llm_load_tensors: CPU buffer size = 315.30 MiB
llm_load_tensors: CUDA0 buffer size = 3127.61 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 = 18.00 MiB
llama_new_context_with_model: KV self size = 18.00 MiB, K (f16): 9.00 MiB, V (f16): 9.00 MiB
llama_new_context_with_model: CUDA_Host output buffer size = 0.58 MiB
llama_new_context_with_model: CUDA0 compute buffer size = 300.75 MiB
llama_new_context_with_model: CUDA_Host compute buffer size = 5.01 MiB
llama_new_context_with_model: graph nodes = 1266
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 | 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 135.302 ms
compute_imatrix: computing over 128 chunks with batch_size 512
compute_imatrix: 0.53 seconds per pass - ETA 1.12 minutes
[1]5.4725,[2]3.7782,[3]3.7538,[4]4.3429,[5]4.1577,[6]3.8350,[7]4.2738,[8]4.3086,[9]4.7510,[10]4.5923,[11]4.5057,[12]4.9433,[13]5.4815,[14]5.7276,[15]6.2856,[16]6.6246,[17]6.8589,[18]7.3081,[19]7.1124,[20]7.2169,[21]7.2418,[22]7.2679,[23]7.1133,[24]7.3331,[25]7.5405,[26]7.4027,[27]7.5308,[28]7.6387,[29]7.8488,[30]7.8116,[31]7.5495,[32]7.2557,[33]7.0906,[34]6.9792,[35]6.8933,[36]6.8737,[37]6.8806,[38]6.9562,[39]6.9103,[40]7.0934,[41]7.1521,[42]7.4350,[43]7.6915,[44]7.9158,[45]8.0726,[46]8.1978,[47]8.0729,[48]8.1199,[49]8.2099,[50]8.2747,[51]8.1531,[52]8.2316,[53]8.3993,[54]8.5016,[55]8.6094,[56]8.6917,[57]8.7388,[58]8.8038,[59]8.8242,[60]8.8550,[61]8.8206,[62]8.7785,[63]8.8293,[64]8.8878,[65]8.8225,[66]8.8215,[67]8.8153,[68]8.7218,[69]8.6522,[70]8.6330,[71]8.5902,[72]8.5736,[73]8.5788,[74]8.4884,[75]8.4141,[76]8.3351,[77]8.3025,[78]8.2680,[79]8.2208,[80]8.1340,[81]8.1606,[82]8.1431,[83]8.0850,[84]8.1047,[85]8.1147,[86]8.0559,[87]8.0283,[88]8.0199,[89]8.0413,[90]8.0703,[91]8.0663,[92]8.0062,[93]7.9531,[94]7.8856,[95]7.8208,[96]7.7639,[97]7.6987,[98]7.6445,[99]7.6172,[100]7.6257,[101]7.6478,[102]7.7493,[103]7.8420,[104]7.9199,[105]8.0490,[106]8.1359,[107]8.1713,[108]8.1466,[109]8.1379,[110]8.1185,[111]8.0864,[112]8.0307,[113]8.0410,[114]8.0888,[115]8.0976,[116]8.1109,[117]8.1307,[118]8.1667,[119]8.1699,[120]8.1621,[121]8.1729,[122]8.1367,[123]8.1824,[124]8.2372,[125]8.2794,[126]8.3367,[127]8.3891,[128]8.4361,
Final estimate: PPL = 8.4361 +/- 0.12256
llama_perf_context_print: load time = 1601.73 ms
llama_perf_context_print: prompt eval time = 48007.13 ms / 65536 tokens ( 0.73 ms per token, 1365.13 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 = 50067.68 ms / 65537 tokens
|