|
llama_model_loader: loaded meta data with 30 key-value pairs and 435 tensors from internlm2_5-20b-chat-IMat-GGUF/internlm2_5-20b-chat.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 = internlm2 |
|
llama_model_loader: - kv 1: general.type str = model |
|
llama_model_loader: - kv 2: general.name str = Internlm2_5 20b Chat |
|
llama_model_loader: - kv 3: general.finetune str = chat |
|
llama_model_loader: - kv 4: general.basename str = internlm2_5 |
|
llama_model_loader: - kv 5: general.size_label str = 20B |
|
llama_model_loader: - kv 6: general.license str = other |
|
llama_model_loader: - kv 7: general.tags arr[str,1] = ["text-generation"] |
|
llama_model_loader: - kv 8: internlm2.context_length u32 = 32768 |
|
llama_model_loader: - kv 9: internlm2.block_count u32 = 48 |
|
llama_model_loader: - kv 10: internlm2.embedding_length u32 = 6144 |
|
llama_model_loader: - kv 11: internlm2.feed_forward_length u32 = 16384 |
|
llama_model_loader: - kv 12: internlm2.rope.freq_base f32 = 50000000.000000 |
|
llama_model_loader: - kv 13: internlm2.attention.head_count u32 = 48 |
|
llama_model_loader: - kv 14: internlm2.attention.layer_norm_rms_epsilon f32 = 0.000010 |
|
llama_model_loader: - kv 15: internlm2.attention.head_count_kv u32 = 8 |
|
llama_model_loader: - kv 16: general.file_type u32 = 7 |
|
llama_model_loader: - kv 17: tokenizer.ggml.model str = llama |
|
llama_model_loader: - kv 18: tokenizer.ggml.pre str = default |
|
llama_model_loader: - kv 19: tokenizer.ggml.tokens arr[str,92544] = ["<unk>", "<s>", "</s>", "<0x00>", "<... |
|
llama_model_loader: - kv 20: tokenizer.ggml.scores arr[f32,92544] = [-1000.000000, -1000.000000, -1000.00... |
|
llama_model_loader: - kv 21: tokenizer.ggml.token_type arr[i32,92544] = [3, 3, 3, 6, 6, 6, 6, 6, 6, 6, 6, 6, ... |
|
llama_model_loader: - kv 22: tokenizer.ggml.add_space_prefix bool = false |
|
llama_model_loader: - kv 23: tokenizer.ggml.bos_token_id u32 = 1 |
|
llama_model_loader: - kv 24: tokenizer.ggml.eos_token_id u32 = 92542 |
|
llama_model_loader: - kv 25: tokenizer.ggml.padding_token_id u32 = 2 |
|
llama_model_loader: - kv 26: tokenizer.ggml.add_bos_token bool = true |
|
llama_model_loader: - kv 27: tokenizer.ggml.add_eos_token bool = false |
|
llama_model_loader: - kv 28: tokenizer.chat_template str = {{ bos_token }}{% for message in mess... |
|
llama_model_loader: - kv 29: general.quantization_version u32 = 2 |
|
llama_model_loader: - type f32: 97 tensors |
|
llama_model_loader: - type q8_0: 338 tensors |
|
llm_load_vocab: special tokens cache size = 9 |
|
llm_load_vocab: token to piece cache size = 0.5508 MB |
|
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: vocab_only = 0 |
|
llm_load_print_meta: n_ctx_train = 32768 |
|
llm_load_print_meta: n_embd = 6144 |
|
llm_load_print_meta: n_layer = 48 |
|
llm_load_print_meta: n_head = 48 |
|
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 = 6 |
|
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 = 16384 |
|
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 = 50000000.0 |
|
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: model type = 20B |
|
llm_load_print_meta: model ftype = Q8_0 |
|
llm_load_print_meta: model params = 19.86 B |
|
llm_load_print_meta: model size = 19.65 GiB (8.50 BPW) |
|
llm_load_print_meta: general.name = Internlm2_5 20b Chat |
|
llm_load_print_meta: BOS token = 1 '<s>' |
|
llm_load_print_meta: EOS token = 92542 '<|im_end|>' |
|
llm_load_print_meta: UNK token = 0 '<unk>' |
|
llm_load_print_meta: PAD token = 2 '</s>' |
|
llm_load_print_meta: LF token = 13 '<0x0A>' |
|
llm_load_print_meta: EOT token = 92542 '<|im_end|>' |
|
llm_load_print_meta: max token length = 384 |
|
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 = 0.41 MiB |
|
llm_load_tensors: offloading 48 repeating layers to GPU |
|
llm_load_tensors: offloading non-repeating layers to GPU |
|
llm_load_tensors: offloaded 49/49 layers to GPU |
|
llm_load_tensors: CPU buffer size = 576.14 MiB |
|
llm_load_tensors: CUDA0 buffer size = 19550.41 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 = 50000000.0 |
|
llama_new_context_with_model: freq_scale = 1 |
|
llama_kv_cache_init: CUDA0 KV buffer size = 96.00 MiB |
|
llama_new_context_with_model: KV self size = 96.00 MiB, K (f16): 48.00 MiB, V (f16): 48.00 MiB |
|
llama_new_context_with_model: CUDA_Host output buffer size = 0.35 MiB |
|
llama_new_context_with_model: CUDA0 compute buffer size = 192.75 MiB |
|
llama_new_context_with_model: CUDA_Host compute buffer size = 13.01 MiB |
|
llama_new_context_with_model: graph nodes = 1542 |
|
llama_new_context_with_model: graph splits = 2 |
|
|
|
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 104.292 ms |
|
compute_imatrix: computing over 136 chunks with batch_size 512 |
|
compute_imatrix: 1.24 seconds per pass - ETA 2.80 minutes |
|
[1]4.9433,[2]3.5465,[3]3.3652,[4]3.8151,[5]3.7627,[6]3.4044,[7]3.9572,[8]3.9796,[9]4.3954, |
|
save_imatrix: stored collected data after 10 chunks in internlm2_5-20b-chat-IMat-GGUF/imatrix.dat |
|
[10]4.5468,[11]4.1983,[12]4.5041,[13]5.0584,[14]5.3707,[15]5.7616,[16]5.9628,[17]5.7050,[18]5.8930,[19]6.1854, |
|
save_imatrix: stored collected data after 20 chunks in internlm2_5-20b-chat-IMat-GGUF/imatrix.dat |
|
[20]5.9909,[21]6.0645,[22]6.1454,[23]6.1814,[24]6.1374,[25]6.2946,[26]6.4994,[27]6.6498,[28]6.6710,[29]6.7504, |
|
save_imatrix: stored collected data after 30 chunks in internlm2_5-20b-chat-IMat-GGUF/imatrix.dat |
|
[30]6.9566,[31]7.0051,[32]6.7743,[33]6.5221,[34]6.2758,[35]6.0577,[36]5.9228,[37]5.8361,[38]5.7738,[39]5.7151, |
|
save_imatrix: stored collected data after 40 chunks in internlm2_5-20b-chat-IMat-GGUF/imatrix.dat |
|
[40]5.6441,[41]5.6000,[42]5.5048,[43]5.4601,[44]5.5103,[45]5.5369,[46]5.6457,[47]5.6336,[48]5.7881,[49]5.8867, |
|
save_imatrix: stored collected data after 50 chunks in internlm2_5-20b-chat-IMat-GGUF/imatrix.dat |
|
[50]6.0185,[51]6.1210,[52]6.2534,[53]6.1524,[54]6.2275,[55]6.2990,[56]6.4089,[57]6.2990,[58]6.3198,[59]6.3431, |
|
save_imatrix: stored collected data after 60 chunks in internlm2_5-20b-chat-IMat-GGUF/imatrix.dat |
|
[60]6.4292,[61]6.5505,[62]6.6334,[63]6.6961,[64]6.7096,[65]6.7299,[66]6.7194,[67]6.6990,[68]6.6319,[69]6.6265, |
|
save_imatrix: stored collected data after 70 chunks in internlm2_5-20b-chat-IMat-GGUF/imatrix.dat |
|
[70]6.6806,[71]6.7003,[72]6.6329,[73]6.5933,[74]6.5994,[75]6.5656,[76]6.5599,[77]6.5486,[78]6.5601,[79]6.5212, |
|
save_imatrix: stored collected data after 80 chunks in internlm2_5-20b-chat-IMat-GGUF/imatrix.dat |
|
[80]6.5208,[81]6.4772,[82]6.4516,[83]6.4219,[84]6.4114,[85]6.3637,[86]6.3351,[87]6.3076,[88]6.3319,[89]6.3471, |
|
save_imatrix: stored collected data after 90 chunks in internlm2_5-20b-chat-IMat-GGUF/imatrix.dat |
|
[90]6.3203,[91]6.3411,[92]6.3604,[93]6.3138,[94]6.3055,[95]6.2836,[96]6.3029,[97]6.3051,[98]6.3081,[99]6.2754, |
|
save_imatrix: stored collected data after 100 chunks in internlm2_5-20b-chat-IMat-GGUF/imatrix.dat |
|
[100]6.2523,[101]6.2147,[102]6.1763,[103]6.1504,[104]6.1218,[105]6.0938,[106]6.0704,[107]6.0809,[108]6.1147,[109]6.1776, |
|
save_imatrix: stored collected data after 110 chunks in internlm2_5-20b-chat-IMat-GGUF/imatrix.dat |
|
[110]6.2404,[111]6.2917,[112]6.3841,[113]6.4398,[114]6.4649,[115]6.4649,[116]6.4770,[117]6.4682,[118]6.4559,[119]6.4013, |
|
save_imatrix: stored collected data after 120 chunks in internlm2_5-20b-chat-IMat-GGUF/imatrix.dat |
|
[120]6.3480,[121]6.3764,[122]6.3763,[123]6.3789,[124]6.3993,[125]6.4365,[126]6.4602,[127]6.4740,[128]6.4975,[129]6.5216, |
|
save_imatrix: stored collected data after 130 chunks in internlm2_5-20b-chat-IMat-GGUF/imatrix.dat |
|
[130]6.4913,[131]6.5093,[132]6.5505,[133]6.5754,[134]6.6208,[135]6.6547,[136]6.6856, |
|
save_imatrix: stored collected data after 136 chunks in internlm2_5-20b-chat-IMat-GGUF/imatrix.dat |
|
|
|
llama_print_timings: load time = 8844.06 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 = 159644.81 ms / 69632 tokens ( 2.29 ms per token, 436.17 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 = 167992.10 ms / 69633 tokens |
|
|
|
Final estimate: PPL = 6.6856 +/- 0.08686 |
|
|
