Upload imatrix.log with huggingface_hub

imatrix.log

@@ -0,0 +1,148 @@
+llama_model_loader: loaded meta data with 26 key-value pairs and 291 tensors from mathstral-7B-v0.1-IMat-GGUF/mathstral-7B-v0.1.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              = llama
+llama_model_loader: - kv   1:                               general.type str              = model
+llama_model_loader: - kv   2:                               general.name str              = mathstral-7B-v0.1
+llama_model_loader: - kv   3:                          llama.block_count u32              = 32
+llama_model_loader: - kv   4:                       llama.context_length u32              = 32768
+llama_model_loader: - kv   5:                     llama.embedding_length u32              = 4096
+llama_model_loader: - kv   6:                  llama.feed_forward_length u32              = 14336
+llama_model_loader: - kv   7:                 llama.attention.head_count u32              = 32
+llama_model_loader: - kv   8:              llama.attention.head_count_kv u32              = 8
+llama_model_loader: - kv   9:                       llama.rope.freq_base f32              = 1000000.000000
+llama_model_loader: - kv  10:     llama.attention.layer_norm_rms_epsilon f32              = 0.000010
+llama_model_loader: - kv  11:                          general.file_type u32              = 7
+llama_model_loader: - kv  12:                           llama.vocab_size u32              = 32768
+llama_model_loader: - kv  13:                 llama.rope.dimension_count u32              = 128
+llama_model_loader: - kv  14:            tokenizer.ggml.add_space_prefix bool             = false
+llama_model_loader: - kv  15:                       tokenizer.ggml.model str              = llama
+llama_model_loader: - kv  16:                         tokenizer.ggml.pre str              = default
+llama_model_loader: - kv  17:                      tokenizer.ggml.tokens arr[str,32768]   = ["<unk>", "<s>", "</s>", "[INST]", "[...
+llama_model_loader: - kv  18:                      tokenizer.ggml.scores arr[f32,32768]   = [-1000.000000, -1000.000000, -1000.00...
+llama_model_loader: - kv  19:                  tokenizer.ggml.token_type arr[i32,32768]   = [3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, ...
+llama_model_loader: - kv  20:                tokenizer.ggml.bos_token_id u32              = 1
+llama_model_loader: - kv  21:                tokenizer.ggml.eos_token_id u32              = 2
+llama_model_loader: - kv  22:            tokenizer.ggml.unknown_token_id u32              = 0
+llama_model_loader: - kv  23:               tokenizer.ggml.add_bos_token bool             = true
+llama_model_loader: - kv  24:               tokenizer.ggml.add_eos_token bool             = false
+llama_model_loader: - kv  25:               general.quantization_version u32              = 2
+llama_model_loader: - type  f32:   65 tensors
+llama_model_loader: - type q8_0:  226 tensors
+llm_load_vocab: special tokens cache size = 771
+llm_load_vocab: token to piece cache size = 0.1732 MB
+llm_load_print_meta: format           = GGUF V3 (latest)
+llm_load_print_meta: arch             = llama
+llm_load_print_meta: vocab type       = SPM
+llm_load_print_meta: n_vocab          = 32768
+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           = 4096
+llm_load_print_meta: n_layer          = 32
+llm_load_print_meta: n_head           = 32
+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            = 4
+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             = 14336
+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  = 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: model type       = 7B
+llm_load_print_meta: model ftype      = Q8_0
+llm_load_print_meta: model params     = 7.25 B
+llm_load_print_meta: model size       = 7.17 GiB (8.50 BPW) 
+llm_load_print_meta: general.name     = mathstral-7B-v0.1
+llm_load_print_meta: BOS token        = 1 '<s>'
+llm_load_print_meta: EOS token        = 2 '</s>'
+llm_load_print_meta: UNK token        = 0 '<unk>'
+llm_load_print_meta: LF token         = 781 '<0x0A>'
+llm_load_print_meta: max token length = 48
+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 =    0.27 MiB
+llm_load_tensors: offloading 32 repeating layers to GPU
+llm_load_tensors: offloading non-repeating layers to GPU
+llm_load_tensors: offloaded 33/33 layers to GPU
+llm_load_tensors:        CPU buffer size =   136.00 MiB
+llm_load_tensors:      CUDA0 buffer size =  7209.02 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 =    64.00 MiB
+llama_new_context_with_model: KV self size  =   64.00 MiB, K (f16):   32.00 MiB, V (f16):   32.00 MiB
+llama_new_context_with_model:  CUDA_Host  output buffer size =     0.12 MiB
+llama_new_context_with_model:      CUDA0 compute buffer size =    81.00 MiB
+llama_new_context_with_model:  CUDA_Host compute buffer size =     9.01 MiB
+llama_new_context_with_model: graph nodes  = 1030
+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 92.115 ms
+compute_imatrix: computing over 148 chunks with batch_size 512
+compute_imatrix: 0.69 seconds per pass - ETA 1.70 minutes
+[1]3.5452,[2]2.8329,[3]3.0262,[4]3.1460,[5]3.5080,[6]3.5694,[7]3.2126,[8]3.6830,[9]3.8318,
+save_imatrix: stored collected data after 10 chunks in mathstral-7B-v0.1-IMat-GGUF/imatrix.dat
+[10]4.1986,[11]4.3073,[12]4.0355,[13]4.2441,[14]4.5271,[15]4.9188,[16]5.0713,[17]5.3272,[18]5.4369,[19]5.5136,
+save_imatrix: stored collected data after 20 chunks in mathstral-7B-v0.1-IMat-GGUF/imatrix.dat
+[20]5.6691,[21]5.6314,[22]5.4412,[23]5.5454,[24]5.5499,[25]5.5891,[26]5.4078,[27]5.6123,[28]5.5037,[29]5.5842,
+save_imatrix: stored collected data after 30 chunks in mathstral-7B-v0.1-IMat-GGUF/imatrix.dat
+[30]5.5128,[31]5.6122,[32]5.7471,[33]5.9049,[34]5.9470,[35]5.8957,[36]5.6761,[37]5.5096,[38]5.3793,[39]5.2930,
+save_imatrix: stored collected data after 40 chunks in mathstral-7B-v0.1-IMat-GGUF/imatrix.dat
+[40]5.2242,[41]5.2013,[42]5.1347,[43]5.0977,[44]5.0190,[45]4.9909,[46]5.0124,[47]5.0766,[48]5.1782,[49]5.2092,
+save_imatrix: stored collected data after 50 chunks in mathstral-7B-v0.1-IMat-GGUF/imatrix.dat
+[50]5.3932,[51]5.5280,[52]5.7135,[53]5.8522,[54]5.9860,[55]5.9293,[56]5.8975,[57]5.9817,[58]6.0532,[59]6.0657,
+save_imatrix: stored collected data after 60 chunks in mathstral-7B-v0.1-IMat-GGUF/imatrix.dat
+[60]6.0003,[61]5.9817,[62]5.9937,[63]6.0510,[64]6.1530,[65]6.2290,[66]6.2539,[67]6.2774,[68]6.3050,[69]6.3105,
+save_imatrix: stored collected data after 70 chunks in mathstral-7B-v0.1-IMat-GGUF/imatrix.dat
+[70]6.3049,[71]6.2421,[72]6.1953,[73]6.1624,[74]6.1771,[75]6.2074,[76]6.1869,[77]6.1996,[78]6.2031,[79]6.1725,
+save_imatrix: stored collected data after 80 chunks in mathstral-7B-v0.1-IMat-GGUF/imatrix.dat
+[80]6.1499,[81]6.1183,[82]6.1181,[83]6.1110,[84]6.1010,[85]6.1118,[86]6.0884,[87]6.0691,[88]6.0443,[89]6.0441,
+save_imatrix: stored collected data after 90 chunks in mathstral-7B-v0.1-IMat-GGUF/imatrix.dat
+[90]6.0145,[91]5.9830,[92]5.9591,[93]5.9387,[94]5.9632,[95]5.9798,[96]5.9557,[97]5.9501,[98]5.9288,[99]5.9606,
+save_imatrix: stored collected data after 100 chunks in mathstral-7B-v0.1-IMat-GGUF/imatrix.dat
+[100]5.9009,[101]5.8985,[102]5.8895,[103]5.9036,[104]5.9168,[105]5.9061,[106]5.8697,[107]5.8317,[108]5.7936,[109]5.7524,
+save_imatrix: stored collected data after 110 chunks in mathstral-7B-v0.1-IMat-GGUF/imatrix.dat
+[110]5.7113,[111]5.6772,[112]5.6418,[113]5.6044,[114]5.5670,[115]5.5359,[116]5.5421,[117]5.5597,[118]5.6181,[119]5.6725,
+save_imatrix: stored collected data after 120 chunks in mathstral-7B-v0.1-IMat-GGUF/imatrix.dat
+[120]5.7221,[121]5.8018,[122]5.8702,[123]5.8742,[124]5.8832,[125]5.8403,[126]5.8329,[127]5.8244,[128]5.8196,[129]5.7959,
+save_imatrix: stored collected data after 130 chunks in mathstral-7B-v0.1-IMat-GGUF/imatrix.dat
+[130]5.7644,[131]5.7874,[132]5.8225,[133]5.8230,[134]5.8249,[135]5.8384,[136]5.8640,[137]5.8734,[138]5.8819,[139]5.9024,
+save_imatrix: stored collected data after 140 chunks in mathstral-7B-v0.1-IMat-GGUF/imatrix.dat
+[140]5.9061,[141]5.8953,[142]5.9452,[143]5.9894,[144]6.0205,[145]6.0665,[146]6.1046,[147]6.1571,[148]6.1966,
+save_imatrix: stored collected data after 148 chunks in mathstral-7B-v0.1-IMat-GGUF/imatrix.dat
+
+llama_print_timings:        load time =    3695.51 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 =   83480.72 ms / 75776 tokens (    1.10 ms per token,   907.71 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 =   86882.14 ms / 75777 tokens
+
+Final estimate: PPL = 6.1966 +/- 0.07627