|
llama_model_loader: loaded meta data with 29 key-value pairs and 290 tensors from reader-lm-0.5b-IMat-GGUF/reader-lm-0.5b.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 0.5b 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 = 0.5B |
|
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 = 24 |
|
llama_model_loader: - kv 11: qwen2.context_length u32 = 256000 |
|
llama_model_loader: - kv 12: qwen2.embedding_length u32 = 896 |
|
llama_model_loader: - kv 13: qwen2.feed_forward_length u32 = 4864 |
|
llama_model_loader: - kv 14: qwen2.attention.head_count u32 = 14 |
|
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: 121 tensors |
|
llama_model_loader: - type q8_0: 169 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 = 896 |
|
llm_load_print_meta: n_layer = 24 |
|
llm_load_print_meta: n_head = 14 |
|
llm_load_print_meta: n_head_kv = 2 |
|
llm_load_print_meta: n_rot = 64 |
|
llm_load_print_meta: n_swa = 0 |
|
llm_load_print_meta: n_embd_head_k = 64 |
|
llm_load_print_meta: n_embd_head_v = 64 |
|
llm_load_print_meta: n_gqa = 7 |
|
llm_load_print_meta: n_embd_k_gqa = 128 |
|
llm_load_print_meta: n_embd_v_gqa = 128 |
|
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 = 4864 |
|
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 = 1B |
|
llm_load_print_meta: model ftype = Q8_0 |
|
llm_load_print_meta: model params = 494.03 M |
|
llm_load_print_meta: model size = 500.79 MiB (8.50 BPW) |
|
llm_load_print_meta: general.name = Qwen2 0.5b 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 = 0.25 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: CPU buffer size = 137.94 MiB |
|
llm_load_tensors: CUDA0 buffer size = 500.84 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 = 6.00 MiB |
|
llama_new_context_with_model: KV self size = 6.00 MiB, K (f16): 3.00 MiB, V (f16): 3.00 MiB |
|
llama_new_context_with_model: CUDA_Host output buffer size = 0.58 MiB |
|
llama_new_context_with_model: CUDA0 compute buffer size = 298.50 MiB |
|
llama_new_context_with_model: CUDA_Host compute buffer size = 2.76 MiB |
|
llama_new_context_with_model: graph nodes = 846 |
|
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 135.191 ms |
|
compute_imatrix: computing over 128 chunks with batch_size 512 |
|
compute_imatrix: 0.31 seconds per pass - ETA 0.67 minutes |
|
[1]17.5855,[2]30.5654,[3]24.3179,[4]23.8257,[5]28.3527,[6]23.2434,[7]23.5053,[8]27.3149,[9]26.8703, |
|
save_imatrix: stored collected data after 10 chunks in reader-lm-0.5b-IMat-GGUF/imatrix.dat |
|
[10]22.9816,[11]20.9238,[12]21.3352,[13]24.0195,[14]25.3831,[15]26.3731,[16]26.5095,[17]25.8473,[18]28.2340,[19]27.2074, |
|
save_imatrix: stored collected data after 20 chunks in reader-lm-0.5b-IMat-GGUF/imatrix.dat |
|
[20]26.4319,[21]26.1346,[22]25.6698,[23]26.5493,[24]28.6411,[25]30.0712,[26]29.5481,[27]31.1370,[28]31.3748,[29]31.8980, |
|
save_imatrix: stored collected data after 30 chunks in reader-lm-0.5b-IMat-GGUF/imatrix.dat |
|
[30]33.4559,[31]32.5650,[32]31.4434,[33]30.9599,[34]30.5894,[35]29.5064,[36]29.1429,[37]30.0905,[38]29.9891,[39]30.7946, |
|
save_imatrix: stored collected data after 40 chunks in reader-lm-0.5b-IMat-GGUF/imatrix.dat |
|
[40]30.7759,[41]31.4073,[42]33.0237,[43]33.7226,[44]34.4842,[45]34.9874,[46]35.1861,[47]34.3317,[48]34.6818,[49]34.6933, |
|
save_imatrix: stored collected data after 50 chunks in reader-lm-0.5b-IMat-GGUF/imatrix.dat |
|
[50]35.2924,[51]34.6092,[52]35.2739,[53]35.5858,[54]35.5820,[55]36.6457,[56]37.5340,[57]37.9605,[58]38.5727,[59]39.0046, |
|
save_imatrix: stored collected data after 60 chunks in reader-lm-0.5b-IMat-GGUF/imatrix.dat |
|
[60]38.4983,[61]37.9428,[62]37.4100,[63]37.1360,[64]38.0599,[65]37.5736,[66]37.8492,[67]38.0620,[68]37.5754,[69]37.2230, |
|
save_imatrix: stored collected data after 70 chunks in reader-lm-0.5b-IMat-GGUF/imatrix.dat |
|
[70]36.9500,[71]36.4919,[72]36.6494,[73]37.1150,[74]36.8300,[75]36.2223,[76]36.1447,[77]36.1864,[78]36.4957,[79]36.8705, |
|
save_imatrix: stored collected data after 80 chunks in reader-lm-0.5b-IMat-GGUF/imatrix.dat |
|
[80]36.4471,[81]36.7004,[82]36.3808,[83]35.8913,[84]36.3684,[85]36.1145,[86]35.6615,[87]35.1692,[88]35.3073,[89]36.0478, |
|
save_imatrix: stored collected data after 90 chunks in reader-lm-0.5b-IMat-GGUF/imatrix.dat |
|
[90]35.8692,[91]36.1951,[92]35.4915,[93]35.2934,[94]34.7298,[95]34.0772,[96]33.5280,[97]33.1758,[98]32.6341,[99]33.0392, |
|
save_imatrix: stored collected data after 100 chunks in reader-lm-0.5b-IMat-GGUF/imatrix.dat |
|
[100]32.8555,[101]32.6738,[102]33.3807,[103]33.5267,[104]33.6360,[105]34.2168,[106]34.2754,[107]34.1549,[108]34.4034,[109]34.5580, |
|
save_imatrix: stored collected data after 110 chunks in reader-lm-0.5b-IMat-GGUF/imatrix.dat |
|
[110]34.4285,[111]34.1185,[112]33.7908,[113]33.6723,[114]33.5806,[115]33.3842,[116]33.6631,[117]33.7834,[118]33.6583,[119]33.6815, |
|
save_imatrix: stored collected data after 120 chunks in reader-lm-0.5b-IMat-GGUF/imatrix.dat |
|
[120]33.3917,[121]33.2004,[122]32.9209,[123]33.1892,[124]33.3031,[125]33.4261,[126]33.8849,[127]34.0284,[128]34.1452, |
|
save_imatrix: stored collected data after 128 chunks in reader-lm-0.5b-IMat-GGUF/imatrix.dat |
|
|
|
llama_perf_print: load time = 716.69 ms |
|
llama_perf_print: prompt eval time = 17498.94 ms / 65536 tokens ( 0.27 ms per token, 3745.14 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 = 18830.03 ms / 65537 tokens |
|
|
|
Final estimate: PPL = 34.1452 +/- 0.62968 |
|
|
