|
llama_model_loader: loaded meta data with 29 key-value pairs and 338 tensors from reader-lm-1.5b-IMat-GGUF/reader-lm-1.5b.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 1.5b Reader |
|
llama_model_loader: - kv 3: general.organization str = Jinaai |
|
llama_model_loader: - kv 4: general.finetune str = reader |
|
llama_model_loader: - kv 5: general.basename str = qwen2 |
|
llama_model_loader: - kv 6: general.size_label str = 1.5B |
|
llama_model_loader: - kv 7: general.license str = cc-by-nc-4.0 |
|
llama_model_loader: - kv 8: general.tags arr[str,1] = ["text-generation"] |
|
llama_model_loader: - kv 9: general.languages arr[str,1] = ["multilingual"] |
|
llama_model_loader: - kv 10: qwen2.block_count u32 = 28 |
|
llama_model_loader: - kv 11: qwen2.context_length u32 = 256000 |
|
llama_model_loader: - kv 12: qwen2.embedding_length u32 = 1536 |
|
llama_model_loader: - kv 13: qwen2.feed_forward_length u32 = 8960 |
|
llama_model_loader: - kv 14: qwen2.attention.head_count u32 = 12 |
|
llama_model_loader: - kv 15: qwen2.attention.head_count_kv u32 = 2 |
|
llama_model_loader: - kv 16: qwen2.rope.freq_base f32 = 2000000.000000 |
|
llama_model_loader: - kv 17: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000001 |
|
llama_model_loader: - kv 18: general.file_type u32 = 7 |
|
llama_model_loader: - kv 19: tokenizer.ggml.model str = gpt2 |
|
llama_model_loader: - kv 20: tokenizer.ggml.pre str = qwen2 |
|
llama_model_loader: - kv 21: tokenizer.ggml.tokens arr[str,151936] = ["!", "\"", "#", "$", "%", "&", "'", ... |
|
llama_model_loader: - kv 22: tokenizer.ggml.token_type arr[i32,151936] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... |
|
llama_model_loader: - kv 23: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... |
|
llama_model_loader: - kv 24: tokenizer.ggml.eos_token_id u32 = 151645 |
|
llama_model_loader: - kv 25: tokenizer.ggml.padding_token_id u32 = 151643 |
|
llama_model_loader: - kv 26: tokenizer.ggml.bos_token_id u32 = 151643 |
|
llama_model_loader: - kv 27: tokenizer.chat_template str = {% for message in messages %}{% if lo... |
|
llama_model_loader: - kv 28: general.quantization_version u32 = 2 |
|
llama_model_loader: - type f32: 141 tensors |
|
llama_model_loader: - type q8_0: 197 tensors |
|
llm_load_vocab: special tokens cache size = 3 |
|
llm_load_vocab: token to piece cache size = 0.9308 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 = 256000 |
|
llm_load_print_meta: n_embd = 1536 |
|
llm_load_print_meta: n_layer = 28 |
|
llm_load_print_meta: n_head = 12 |
|
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 = 6 |
|
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 = 8960 |
|
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 = 2000000.0 |
|
llm_load_print_meta: freq_scale_train = 1 |
|
llm_load_print_meta: n_ctx_orig_yarn = 256000 |
|
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 = 1.54 B |
|
llm_load_print_meta: model size = 1.53 GiB (8.50 BPW) |
|
llm_load_print_meta: general.name = Qwen2 1.5b Reader |
|
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 = 0.30 MiB |
|
llm_load_tensors: offloading 28 repeating layers to GPU |
|
llm_load_tensors: offloading non-repeating layers to GPU |
|
llm_load_tensors: offloaded 29/29 layers to GPU |
|
llm_load_tensors: CPU buffer size = 236.47 MiB |
|
llm_load_tensors: CUDA0 buffer size = 1564.63 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 = 2000000.0 |
|
llama_new_context_with_model: freq_scale = 1 |
|
llama_kv_cache_init: CUDA0 KV buffer size = 14.00 MiB |
|
llama_new_context_with_model: KV self size = 14.00 MiB, K (f16): 7.00 MiB, V (f16): 7.00 MiB |
|
llama_new_context_with_model: CUDA_Host output buffer size = 0.58 MiB |
|
llama_new_context_with_model: CUDA0 compute buffer size = 299.75 MiB |
|
llama_new_context_with_model: CUDA_Host compute buffer size = 4.01 MiB |
|
llama_new_context_with_model: graph nodes = 986 |
|
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 128.746 ms |
|
compute_imatrix: computing over 128 chunks with batch_size 512 |
|
compute_imatrix: 0.38 seconds per pass - ETA 0.82 minutes |
|
[1]9.2771,[2]6.8644,[3]6.4782,[4]7.4881,[5]7.2917,[6]6.6791,[7]7.3505,[8]7.7499,[9]8.4398, |
|
save_imatrix: stored collected data after 10 chunks in reader-lm-1.5b-IMat-GGUF/imatrix.dat |
|
[10]7.9031,[11]7.7003,[12]8.2857,[13]9.0872,[14]9.3184,[15]10.0522,[16]10.4603,[17]10.5844,[18]11.0963,[19]10.6325, |
|
save_imatrix: stored collected data after 20 chunks in reader-lm-1.5b-IMat-GGUF/imatrix.dat |
|
[20]10.6879,[21]10.7507,[22]10.7492,[23]10.6613,[24]10.9654,[25]11.0832,[26]11.0748,[27]11.4078,[28]11.6878,[29]12.1242, |
|
save_imatrix: stored collected data after 30 chunks in reader-lm-1.5b-IMat-GGUF/imatrix.dat |
|
[30]12.0099,[31]11.7409,[32]11.2796,[33]10.9901,[34]10.8007,[35]10.5697,[36]10.5644,[37]10.6097,[38]10.7267,[39]10.7439, |
|
save_imatrix: stored collected data after 40 chunks in reader-lm-1.5b-IMat-GGUF/imatrix.dat |
|
[40]10.9118,[41]10.9510,[42]11.4185,[43]11.8320,[44]12.2346,[45]12.5533,[46]12.7154,[47]12.5249,[48]12.5437,[49]12.6274, |
|
save_imatrix: stored collected data after 50 chunks in reader-lm-1.5b-IMat-GGUF/imatrix.dat |
|
[50]12.6882,[51]12.5225,[52]12.6191,[53]12.8579,[54]12.9677,[55]13.1179,[56]13.1808,[57]13.1997,[58]13.2493,[59]13.2682, |
|
save_imatrix: stored collected data after 60 chunks in reader-lm-1.5b-IMat-GGUF/imatrix.dat |
|
[60]13.2639,[61]13.1555,[62]13.0986,[63]13.1404,[64]13.2389,[65]13.1620,[66]13.1591,[67]13.1465,[68]13.0630,[69]13.0219, |
|
save_imatrix: stored collected data after 70 chunks in reader-lm-1.5b-IMat-GGUF/imatrix.dat |
|
[70]12.9859,[71]12.9325,[72]12.9059,[73]12.9107,[74]12.7811,[75]12.6892,[76]12.5848,[77]12.5441,[78]12.5483,[79]12.5050, |
|
save_imatrix: stored collected data after 80 chunks in reader-lm-1.5b-IMat-GGUF/imatrix.dat |
|
[80]12.4548,[81]12.4615,[82]12.4451,[83]12.3640,[84]12.4103,[85]12.4209,[86]12.3428,[87]12.2794,[88]12.2333,[89]12.2637, |
|
save_imatrix: stored collected data after 90 chunks in reader-lm-1.5b-IMat-GGUF/imatrix.dat |
|
[90]12.2632,[91]12.2392,[92]12.1162,[93]12.0106,[94]11.8707,[95]11.7481,[96]11.6479,[97]11.5368,[98]11.4347,[99]11.3816, |
|
save_imatrix: stored collected data after 100 chunks in reader-lm-1.5b-IMat-GGUF/imatrix.dat |
|
[100]11.3768,[101]11.3839,[102]11.5189,[103]11.6346,[104]11.7308,[105]11.9020,[106]11.9952,[107]12.0287,[108]11.9878,[109]11.9411, |
|
save_imatrix: stored collected data after 110 chunks in reader-lm-1.5b-IMat-GGUF/imatrix.dat |
|
[110]11.9535,[111]11.9408,[112]11.8850,[113]11.9203,[114]11.9687,[115]11.9471,[116]11.9409,[117]11.9471,[118]11.9818,[119]11.9618, |
|
save_imatrix: stored collected data after 120 chunks in reader-lm-1.5b-IMat-GGUF/imatrix.dat |
|
[120]11.9384,[121]11.9241,[122]11.8738,[123]11.9322,[124]12.0272,[125]12.1018,[126]12.1950,[127]12.2749,[128]12.3603, |
|
save_imatrix: stored collected data after 128 chunks in reader-lm-1.5b-IMat-GGUF/imatrix.dat |
|
|
|
llama_perf_print: load time = 926.91 ms |
|
llama_perf_print: prompt eval time = 30274.30 ms / 65536 tokens ( 0.46 ms per token, 2164.74 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 = 31809.38 ms / 65537 tokens |
|
|
|
Final estimate: PPL = 12.3603 +/- 0.18881 |
|
|
