legraphista's picture
Upload imatrix.log with huggingface_hub
6231ed5 verified
raw
history blame
12.4 kB
llama_model_loader: loaded meta data with 38 key-value pairs and 377 tensors from DeepSeek-Coder-V2-Lite-Instruct-IMat-GGUF/DeepSeek-Coder-V2-Lite-Instruct.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 = deepseek2
llama_model_loader: - kv 1: general.name str = DeepSeek-Coder-V2-Lite-Instruct
llama_model_loader: - kv 2: deepseek2.block_count u32 = 27
llama_model_loader: - kv 3: deepseek2.context_length u32 = 163840
llama_model_loader: - kv 4: deepseek2.embedding_length u32 = 2048
llama_model_loader: - kv 5: deepseek2.feed_forward_length u32 = 10944
llama_model_loader: - kv 6: deepseek2.attention.head_count u32 = 16
llama_model_loader: - kv 7: deepseek2.attention.head_count_kv u32 = 16
llama_model_loader: - kv 8: deepseek2.rope.freq_base f32 = 10000.000000
llama_model_loader: - kv 9: deepseek2.attention.layer_norm_rms_epsilon f32 = 0.000001
llama_model_loader: - kv 10: deepseek2.expert_used_count u32 = 6
llama_model_loader: - kv 11: general.file_type u32 = 0
llama_model_loader: - kv 12: deepseek2.leading_dense_block_count u32 = 1
llama_model_loader: - kv 13: deepseek2.vocab_size u32 = 102400
llama_model_loader: - kv 14: deepseek2.attention.kv_lora_rank u32 = 512
llama_model_loader: - kv 15: deepseek2.attention.key_length u32 = 192
llama_model_loader: - kv 16: deepseek2.attention.value_length u32 = 128
llama_model_loader: - kv 17: deepseek2.expert_feed_forward_length u32 = 1408
llama_model_loader: - kv 18: deepseek2.expert_count u32 = 64
llama_model_loader: - kv 19: deepseek2.expert_shared_count u32 = 2
llama_model_loader: - kv 20: deepseek2.expert_weights_scale f32 = 1.000000
llama_model_loader: - kv 21: deepseek2.rope.dimension_count u32 = 64
llama_model_loader: - kv 22: deepseek2.rope.scaling.type str = yarn
llama_model_loader: - kv 23: deepseek2.rope.scaling.factor f32 = 40.000000
llama_model_loader: - kv 24: deepseek2.rope.scaling.original_context_length u32 = 4096
llama_model_loader: - kv 25: deepseek2.rope.scaling.yarn_log_multiplier f32 = 0.070700
llama_model_loader: - kv 26: tokenizer.ggml.model str = gpt2
llama_model_loader: - kv 27: tokenizer.ggml.pre str = deepseek-llm
llama_model_loader: - kv 28: tokenizer.ggml.tokens arr[str,102400] = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv 29: tokenizer.ggml.token_type arr[i32,102400] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv 30: tokenizer.ggml.merges arr[str,99757] = ["Ġ Ġ", "Ġ t", "Ġ a", "i n", "h e...
llama_model_loader: - kv 31: tokenizer.ggml.bos_token_id u32 = 100000
llama_model_loader: - kv 32: tokenizer.ggml.eos_token_id u32 = 100001
llama_model_loader: - kv 33: tokenizer.ggml.padding_token_id u32 = 100001
llama_model_loader: - kv 34: tokenizer.ggml.add_bos_token bool = true
llama_model_loader: - kv 35: tokenizer.ggml.add_eos_token bool = false
llama_model_loader: - kv 36: tokenizer.chat_template str = {% if not add_generation_prompt is de...
llama_model_loader: - kv 37: general.quantization_version u32 = 2
llama_model_loader: - type f32: 377 tensors
llm_load_vocab: special tokens cache size = 2400
llm_load_vocab: token to piece cache size = 0.6661 MB
llm_load_print_meta: format = GGUF V3 (latest)
llm_load_print_meta: arch = deepseek2
llm_load_print_meta: vocab type = BPE
llm_load_print_meta: n_vocab = 102400
llm_load_print_meta: n_merges = 99757
llm_load_print_meta: n_ctx_train = 163840
llm_load_print_meta: n_embd = 2048
llm_load_print_meta: n_head = 16
llm_load_print_meta: n_head_kv = 16
llm_load_print_meta: n_layer = 27
llm_load_print_meta: n_rot = 64
llm_load_print_meta: n_embd_head_k = 192
llm_load_print_meta: n_embd_head_v = 128
llm_load_print_meta: n_gqa = 1
llm_load_print_meta: n_embd_k_gqa = 3072
llm_load_print_meta: n_embd_v_gqa = 2048
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 = 10944
llm_load_print_meta: n_expert = 64
llm_load_print_meta: n_expert_used = 6
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 = yarn
llm_load_print_meta: freq_base_train = 10000.0
llm_load_print_meta: freq_scale_train = 0.025
llm_load_print_meta: n_ctx_orig_yarn = 4096
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 = 16B
llm_load_print_meta: model ftype = all F32
llm_load_print_meta: model params = 15.71 B
llm_load_print_meta: model size = 58.51 GiB (32.00 BPW)
llm_load_print_meta: general.name = DeepSeek-Coder-V2-Lite-Instruct
llm_load_print_meta: BOS token = 100000 '<|begin▁of▁sentence|>'
llm_load_print_meta: EOS token = 100001 '<|end▁of▁sentence|>'
llm_load_print_meta: PAD token = 100001 '<|end▁of▁sentence|>'
llm_load_print_meta: LF token = 126 'Ä'
llm_load_print_meta: n_layer_dense_lead = 1
llm_load_print_meta: n_lora_q = 0
llm_load_print_meta: n_lora_kv = 512
llm_load_print_meta: n_ff_exp = 1408
llm_load_print_meta: n_expert_shared = 2
llm_load_print_meta: expert_weights_scale = 1.0
llm_load_print_meta: rope_yarn_log_mul = 0.0707
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 = 0.35 MiB
llm_load_tensors: offloading 10 repeating layers to GPU
llm_load_tensors: offloaded 10/28 layers to GPU
llm_load_tensors: CPU buffer size = 37605.31 MiB
llm_load_tensors: CUDA0 buffer size = 22310.18 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 = 10000.0
llama_new_context_with_model: freq_scale = 0.025
llama_kv_cache_init: CUDA_Host KV buffer size = 85.00 MiB
llama_kv_cache_init: CUDA0 KV buffer size = 50.00 MiB
llama_new_context_with_model: KV self size = 135.00 MiB, K (f16): 81.00 MiB, V (f16): 54.00 MiB
llama_new_context_with_model: CUDA_Host output buffer size = 0.39 MiB
llama_new_context_with_model: CUDA0 compute buffer size = 1012.00 MiB
llama_new_context_with_model: CUDA_Host compute buffer size = 11.01 MiB
llama_new_context_with_model: graph nodes = 1924
llama_new_context_with_model: graph splits = 272
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 215.372 ms
compute_imatrix: computing over 139 chunks with batch_size 512
compute_imatrix: 2.97 seconds per pass - ETA 6.88 minutes
[1]8.3398,[2]5.4215,[3]5.2838,[4]6.0526,[5]5.6694,[6]5.3675,[7]5.7148,[8]5.8289,[9]6.5437,
save_imatrix: stored collected data after 10 chunks in DeepSeek-Coder-V2-Lite-Instruct-IMat-GGUF/imatrix.dat
[10]6.6919,[11]6.1685,[12]6.4804,[13]6.9224,[14]7.2573,[15]7.3708,[16]7.7938,[17]8.0288,[18]8.1908,[19]8.5073,
save_imatrix: stored collected data after 20 chunks in DeepSeek-Coder-V2-Lite-Instruct-IMat-GGUF/imatrix.dat
[20]8.0227,[21]7.9922,[22]7.9696,[23]8.1526,[24]7.9689,[25]8.2193,[26]8.0970,[27]8.2980,[28]8.1099,[29]8.3295,
save_imatrix: stored collected data after 30 chunks in DeepSeek-Coder-V2-Lite-Instruct-IMat-GGUF/imatrix.dat
[30]8.5506,[31]8.5974,[32]8.4411,[33]8.1166,[34]7.6767,[35]7.3223,[36]7.2017,[37]7.0882,[38]7.0476,[39]6.9338,
save_imatrix: stored collected data after 40 chunks in DeepSeek-Coder-V2-Lite-Instruct-IMat-GGUF/imatrix.dat
[40]6.8448,[41]6.6842,[42]6.5776,[43]6.6411,[44]6.7147,[45]6.8307,[46]6.8212,[47]7.0831,[48]7.3037,[49]7.4481,
save_imatrix: stored collected data after 50 chunks in DeepSeek-Coder-V2-Lite-Instruct-IMat-GGUF/imatrix.dat
[50]7.5775,[51]7.6764,[52]7.5752,[53]7.7096,[54]7.8061,[55]7.9151,[56]7.8063,[57]7.7795,[58]7.7842,[59]7.8984,
save_imatrix: stored collected data after 60 chunks in DeepSeek-Coder-V2-Lite-Instruct-IMat-GGUF/imatrix.dat
[60]8.0463,[61]8.1568,[62]8.2185,[63]8.2355,[64]8.2746,[65]8.2711,[66]8.2385,[67]8.2192,[68]8.1754,[69]8.2307,
save_imatrix: stored collected data after 70 chunks in DeepSeek-Coder-V2-Lite-Instruct-IMat-GGUF/imatrix.dat
[70]8.2569,[71]8.2581,[72]8.2878,[73]8.2914,[74]8.2632,[75]8.2405,[76]8.2185,[77]8.2144,[78]8.2302,[79]8.1705,
save_imatrix: stored collected data after 80 chunks in DeepSeek-Coder-V2-Lite-Instruct-IMat-GGUF/imatrix.dat
[80]8.2038,[81]8.1894,[82]8.1776,[83]8.1431,[84]8.1184,[85]8.0975,[86]8.0670,[87]8.0352,[88]8.0625,[89]8.0887,
save_imatrix: stored collected data after 90 chunks in DeepSeek-Coder-V2-Lite-Instruct-IMat-GGUF/imatrix.dat
[90]8.0910,[91]8.1317,[92]8.1702,[93]8.1019,[94]8.0817,[95]8.0496,[96]8.0788,[97]8.0812,[98]8.0650,[99]7.9677,
save_imatrix: stored collected data after 100 chunks in DeepSeek-Coder-V2-Lite-Instruct-IMat-GGUF/imatrix.dat
[100]7.8685,[101]7.7737,[102]7.6755,[103]7.5767,[104]7.4886,[105]7.4023,[106]7.3171,[107]7.2332,[108]7.1950,[109]7.2108,
save_imatrix: stored collected data after 110 chunks in DeepSeek-Coder-V2-Lite-Instruct-IMat-GGUF/imatrix.dat
[110]7.2567,[111]7.3272,[112]7.3958,[113]7.4487,[114]7.5589,[115]7.6035,[116]7.6316,[117]7.6348,[118]7.6765,[119]7.6743,
save_imatrix: stored collected data after 120 chunks in DeepSeek-Coder-V2-Lite-Instruct-IMat-GGUF/imatrix.dat
[120]7.6614,[121]7.5726,[122]7.4906,[123]7.5336,[124]7.5893,[125]7.5868,[126]7.5893,[127]7.5967,[128]7.6244,[129]7.6239,
save_imatrix: stored collected data after 130 chunks in DeepSeek-Coder-V2-Lite-Instruct-IMat-GGUF/imatrix.dat
[130]7.6332,[131]7.6606,[132]7.6577,[133]7.6290,[134]7.6764,[135]7.7304,[136]7.7647,[137]7.8148,[138]7.8820,[139]7.9125,
save_imatrix: stored collected data after 139 chunks in DeepSeek-Coder-V2-Lite-Instruct-IMat-GGUF/imatrix.dat
llama_print_timings: load time = 17517.65 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 = 385218.87 ms / 71168 tokens ( 5.41 ms per token, 184.75 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 = 400724.75 ms / 71169 tokens
Final estimate: PPL = 7.9125 +/- 0.12345