llama_model_loader: loaded meta data with 23 key-value pairs and 291 tensors from Llama-3-Instruct-8B-SPPO-Iter3-IMat-GGUF/Llama-3-Instruct-8B-SPPO-Iter3.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.name str = Llama-3-Instruct-8B-SPPO-Iter3 llama_model_loader: - kv 2: llama.block_count u32 = 32 llama_model_loader: - kv 3: llama.context_length u32 = 8192 llama_model_loader: - kv 4: llama.embedding_length u32 = 4096 llama_model_loader: - kv 5: llama.feed_forward_length u32 = 14336 llama_model_loader: - kv 6: llama.attention.head_count u32 = 32 llama_model_loader: - kv 7: llama.attention.head_count_kv u32 = 8 llama_model_loader: - kv 8: llama.rope.freq_base f32 = 500000.000000 llama_model_loader: - kv 9: llama.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 10: general.file_type u32 = 7 llama_model_loader: - kv 11: llama.vocab_size u32 = 128256 llama_model_loader: - kv 12: llama.rope.dimension_count u32 = 128 llama_model_loader: - kv 13: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 14: tokenizer.ggml.pre str = llama-bpe llama_model_loader: - kv 15: tokenizer.ggml.tokens arr[str,128256] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 16: tokenizer.ggml.token_type arr[i32,128256] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 17: tokenizer.ggml.merges arr[str,280147] = ["Ġ Ġ", "Ġ ĠĠĠ", "ĠĠ ĠĠ", "... llama_model_loader: - kv 18: tokenizer.ggml.bos_token_id u32 = 128000 llama_model_loader: - kv 19: tokenizer.ggml.eos_token_id u32 = 128009 llama_model_loader: - kv 20: tokenizer.ggml.padding_token_id u32 = 128009 llama_model_loader: - kv 21: tokenizer.chat_template str = {% set loop_messages = messages %}{% ... llama_model_loader: - kv 22: 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 = 256 llm_load_vocab: token to piece cache size = 0.8000 MB llm_load_print_meta: format = GGUF V3 (latest) llm_load_print_meta: arch = llama llm_load_print_meta: vocab type = BPE llm_load_print_meta: n_vocab = 128256 llm_load_print_meta: n_merges = 280147 llm_load_print_meta: n_ctx_train = 8192 llm_load_print_meta: n_embd = 4096 llm_load_print_meta: n_head = 32 llm_load_print_meta: n_head_kv = 8 llm_load_print_meta: n_layer = 32 llm_load_print_meta: n_rot = 128 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 = 500000.0 llm_load_print_meta: freq_scale_train = 1 llm_load_print_meta: n_ctx_orig_yarn = 8192 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 = 8B llm_load_print_meta: model ftype = Q8_0 llm_load_print_meta: model params = 8.03 B llm_load_print_meta: model size = 7.95 GiB (8.50 BPW) llm_load_print_meta: general.name = Llama-3-Instruct-8B-SPPO-Iter3 llm_load_print_meta: BOS token = 128000 '<|begin_of_text|>' llm_load_print_meta: EOS token = 128009 '<|eot_id|>' llm_load_print_meta: PAD token = 128009 '<|eot_id|>' llm_load_print_meta: LF token = 128 'Ä' llm_load_print_meta: EOT token = 128009 '<|eot_id|>' 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 = 0.27 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 = 532.31 MiB llm_load_tensors: CUDA0 buffer size = 7605.33 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 = 500000.0 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 = 0.49 MiB llama_new_context_with_model: CUDA0 compute buffer size = 258.50 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 64.595 ms compute_imatrix: computing over 125 chunks with batch_size 512 compute_imatrix: 0.83 seconds per pass - ETA 1.72 minutes [1]7.1146,[2]5.5633,[3]4.9033,[4]6.2338,[5]6.4395,[6]5.2909,[7]5.6433,[8]6.2173,[9]6.4763, save_imatrix: stored collected data after 10 chunks in Llama-3-Instruct-8B-SPPO-Iter3-IMat-GGUF/imatrix.dat [10]5.8334,[11]6.3363,[12]6.9164,[13]7.3915,[14]7.8408,[15]8.1553,[16]8.4420,[17]8.6336,[18]8.3251,[19]7.8869, save_imatrix: stored collected data after 20 chunks in Llama-3-Instruct-8B-SPPO-Iter3-IMat-GGUF/imatrix.dat [20]7.8945,[21]8.0643,[22]7.9991,[23]8.3410,[24]8.3056,[25]8.6923,[26]8.7143,[27]8.7966,[28]9.0155,[29]9.0275, save_imatrix: stored collected data after 30 chunks in Llama-3-Instruct-8B-SPPO-Iter3-IMat-GGUF/imatrix.dat [30]9.0186,[31]8.5162,[32]8.0651,[33]7.8415,[34]7.6520,[35]7.7247,[36]7.8072,[37]7.7331,[38]7.8325,[39]8.0203, save_imatrix: stored collected data after 40 chunks in Llama-3-Instruct-8B-SPPO-Iter3-IMat-GGUF/imatrix.dat [40]8.1211,[41]8.2076,[42]8.3124,[43]8.5445,[44]8.6209,[45]8.7726,[46]8.6312,[47]8.7817,[48]8.8636,[49]8.9627, save_imatrix: stored collected data after 50 chunks in Llama-3-Instruct-8B-SPPO-Iter3-IMat-GGUF/imatrix.dat [50]8.8179,[51]8.9467,[52]9.0850,[53]9.1818,[54]9.2510,[55]9.3461,[56]9.3836,[57]9.4482,[58]9.4609,[59]9.4704, save_imatrix: stored collected data after 60 chunks in Llama-3-Instruct-8B-SPPO-Iter3-IMat-GGUF/imatrix.dat [60]9.4077,[61]9.3889,[62]9.4296,[63]9.4772,[64]9.3666,[65]9.3270,[66]9.3448,[67]9.3176,[68]9.2975,[69]9.2811, save_imatrix: stored collected data after 70 chunks in Llama-3-Instruct-8B-SPPO-Iter3-IMat-GGUF/imatrix.dat [70]9.2812,[71]9.2683,[72]9.2650,[73]9.2262,[74]9.1612,[75]9.1591,[76]9.1567,[77]9.1178,[78]9.1121,[79]9.1597, save_imatrix: stored collected data after 80 chunks in Llama-3-Instruct-8B-SPPO-Iter3-IMat-GGUF/imatrix.dat [80]9.2015,[81]9.1945,[82]9.1915,[83]9.2246,[84]9.0902,[85]9.0730,[86]9.0704,[87]9.0823,[88]9.1088,[89]9.1139, save_imatrix: stored collected data after 90 chunks in Llama-3-Instruct-8B-SPPO-Iter3-IMat-GGUF/imatrix.dat [90]9.0465,[91]8.9625,[92]8.8767,[93]8.8114,[94]8.7455,[95]8.6825,[96]8.6344,[97]8.6486,[98]8.6937,[99]8.7894, save_imatrix: stored collected data after 100 chunks in Llama-3-Instruct-8B-SPPO-Iter3-IMat-GGUF/imatrix.dat [100]8.8818,[101]8.9448,[102]9.0881,[103]9.1199,[104]9.1685,[105]9.0797,[106]9.0848,[107]9.0180,[108]8.9430,[109]8.8576, save_imatrix: stored collected data after 110 chunks in Llama-3-Instruct-8B-SPPO-Iter3-IMat-GGUF/imatrix.dat [110]8.9121,[111]8.9707,[112]8.9816,[113]8.9868,[114]9.0345,[115]9.0761,[116]9.0970,[117]9.1362,[118]9.1786,[119]9.1072, save_imatrix: stored collected data after 120 chunks in Llama-3-Instruct-8B-SPPO-Iter3-IMat-GGUF/imatrix.dat [120]9.1169,[121]9.1396,[122]9.1697,[123]9.2171,[124]9.2448,[125]9.2795, save_imatrix: stored collected data after 125 chunks in Llama-3-Instruct-8B-SPPO-Iter3-IMat-GGUF/imatrix.dat llama_print_timings: load time = 2383.46 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 = 87673.99 ms / 64000 tokens ( 1.37 ms per token, 729.98 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 = 90337.25 ms / 64001 tokens Final estimate: PPL = 9.2795 +/- 0.14689