[1708357169] Log start [1708357169] Cmd: /main -m BioMistral-7B_Q4_K_M.gguf -p "What is a Large Language Model?" -n 512 --temp 1 [1708357169] main: build = 0 (unknown) [1708357169] main: built with cc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0 for x86_64-linux-gnu [1708357169] main: seed = 1708357169 [1708357169] main: llama backend init [1708357169] main: load the model and apply lora adapter, if any [1708357169] llama_model_loader: loaded meta data with 21 key-value pairs and 291 tensors from BioMistral-7B_Q4_K_M.gguf (version GGUF V3 (latest)) [1708357169] llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. [1708357169] llama_model_loader: - kv 0: general.architecture str = llama [1708357169] llama_model_loader: - kv 1: general.name str = workspace [1708357169] llama_model_loader: - kv 2: llama.context_length u32 = 32768 [1708357169] llama_model_loader: - kv 3: llama.embedding_length u32 = 4096 [1708357169] llama_model_loader: - kv 4: llama.block_count u32 = 32 [1708357169] llama_model_loader: - kv 5: llama.feed_forward_length u32 = 14336 [1708357169] llama_model_loader: - kv 6: llama.rope.dimension_count u32 = 128 [1708357169] llama_model_loader: - kv 7: llama.attention.head_count u32 = 32 [1708357169] llama_model_loader: - kv 8: llama.attention.head_count_kv u32 = 8 [1708357169] llama_model_loader: - kv 9: llama.attention.layer_norm_rms_epsilon f32 = 0.000010 [1708357169] llama_model_loader: - kv 10: llama.rope.freq_base f32 = 10000.000000 [1708357169] llama_model_loader: - kv 11: general.file_type u32 = 15 [1708357169] llama_model_loader: - kv 12: tokenizer.ggml.model str = llama [1708357169] llama_model_loader: - kv 13: tokenizer.ggml.tokens arr[str,32000] = ["", "~~", "~~", "<0x00>", "<... [1708357169] llama_model_loader: - kv 14: tokenizer.ggml.scores arr[f32,32000] = [-1000.000000, -1000.000000, -1000.00... [1708357169] llama_model_loader: - kv 15: tokenizer.ggml.token_type arr[i32,32000] = [3, 3, 3, 6, 6, 6, 6, 6, 6, 6, 6, 6, ... [1708357169] llama_model_loader: - kv 16: tokenizer.ggml.bos_token_id u32 = 1 [1708357169] llama_model_loader: - kv 17: tokenizer.ggml.eos_token_id u32 = 2 [1708357169] llama_model_loader: - kv 18: tokenizer.ggml.unknown_token_id u32 = 0 [1708357169] llama_model_loader: - kv 19: tokenizer.chat_template str = {{ bos_token }}{% for message in mess... [1708357169] llama_model_loader: - kv 20: general.quantization_version u32 = 2 [1708357169] llama_model_loader: - type f32: 65 tensors [1708357169] llama_model_loader: - type q4_K: 193 tensors [1708357169] llama_model_loader: - type q6_K: 33 tensors [1708357169] llm_load_vocab: special tokens definition check successful ( 259/32000 ). [1708357169] llm_load_print_meta: format = GGUF V3 (latest) [1708357169] llm_load_print_meta: arch = llama [1708357169] llm_load_print_meta: vocab type = SPM [1708357169] llm_load_print_meta: n_vocab = 32000 [1708357169] llm_load_print_meta: n_merges = 0 [1708357169] llm_load_print_meta: n_ctx_train = 32768 [1708357169] llm_load_print_meta: n_embd = 4096 [1708357169] llm_load_print_meta: n_head = 32 [1708357169] llm_load_print_meta: n_head_kv = 8 [1708357169] llm_load_print_meta: n_layer = 32 [1708357169] llm_load_print_meta: n_rot = 128 [1708357169] llm_load_print_meta: n_embd_head_k = 128 [1708357169] llm_load_print_meta: n_embd_head_v = 128 [1708357169] llm_load_print_meta: n_gqa = 4 [1708357169] llm_load_print_meta: n_embd_k_gqa = 1024 [1708357169] llm_load_print_meta: n_embd_v_gqa = 1024 [1708357169] llm_load_print_meta: f_norm_eps = 0.0e+00 [1708357169] llm_load_print_meta: f_norm_rms_eps = 1.0e-05 [1708357169] llm_load_print_meta: f_clamp_kqv = 0.0e+00 [1708357169] llm_load_print_meta: f_max_alibi_bias = 0.0e+00 [1708357169] llm_load_print_meta: n_ff = 14336 [1708357169] llm_load_print_meta: n_expert = 0 [1708357169] llm_load_print_meta: n_expert_used = 0 [1708357169] llm_load_print_meta: rope scaling = linear [1708357169] llm_load_print_meta: freq_base_train = 10000.0 [1708357169] llm_load_print_meta: freq_scale_train = 1 [1708357169] llm_load_print_meta: n_yarn_orig_ctx = 32768 [1708357169] llm_load_print_meta: rope_finetuned = unknown [1708357169] llm_load_print_meta: model type = 7B [1708357169] llm_load_print_meta: model ftype = Q4_K - Medium [1708357169] llm_load_print_meta: model params = 7.24 B [1708357169] llm_load_print_meta: model size = 4.07 GiB (4.83 BPW) [1708357169] llm_load_print_meta: general.name = workspace [1708357169] llm_load_print_meta: BOS token = 1 '~~' [1708357169] llm_load_print_meta: EOS token = 2 '~~' [1708357169] llm_load_print_meta: UNK token = 0 '' [1708357169] llm_load_print_meta: LF token = 13 '<0x0A>' [1708357169] llm_load_tensors: ggml ctx size = 0.11 MiB [1708357178] llm_load_tensors: CPU buffer size = 4165.37 MiB [1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] .[1708357178] [1708357178] llama_new_context_with_model: n_ctx = 512 [1708357178] llama_new_context_with_model: freq_base = 10000.0 [1708357178] llama_new_context_with_model: freq_scale = 1 [1708357178] llama_kv_cache_init: CPU KV buffer size = 64.00 MiB [1708357178] llama_new_context_with_model: KV self size = 64.00 MiB, K (f16): 32.00 MiB, V (f16): 32.00 MiB [1708357178] llama_new_context_with_model: CPU input buffer size = 10.01 MiB [1708357178] llama_new_context_with_model: CPU compute buffer size = 72.00 MiB [1708357178] llama_new_context_with_model: graph splits (measure): 1 [1708357178] warming up the model with an empty run [1708357178] n_ctx: 512 [1708357178] [1708357178] system_info: n_threads = 16 / 32 | AVX = 1 | AVX_VNNI = 0 | AVX2 = 1 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | FMA = 1 | NEON = 0 | ARM_FMA = 0 | F16C = 1 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 0 | SSE3 = 1 | SSSE3 = 1 | VSX = 0 | MATMUL_INT8 = 0 | [1708357178] add_bos: 1 [1708357178] tokenize the prompt [1708357178] prompt: "What is a Large Language Model?" [1708357178] tokens: [ '':1, ' What':1824, ' is':349, ' a':264, ' Large':23292, ' Lang':13550, 'ua':3772, 'ge':490, ' Model':8871, '?':28804 ] [1708357178] recalculate the cached logits (check): embd_inp.empty() false, n_matching_session_tokens 0, embd_inp.size() 10, session_tokens.size() 0, embd_inp.size() 10 [1708357178] inp_pfx: [ '':1, ' ':28705, '':13, '':13, '###':27332, ' Inst':3133, 'ruction':3112, ':':28747, '':13, '':13 ] [1708357178] inp_sfx: [ ' ':28705, '':13, '':13, '###':27332, ' Response':12107, ':':28747, '':13, '':13 ] [1708357178] cml_pfx: [ '':1, ' ':28705, '':13, '<':28789, '|':28766, 'im':321, '_':28730, 'start':2521, '|':28766, '>':28767, 'user':1838, '':13 ] [1708357178] cml_sfx: [ ' <':523, '|':28766, 'im':321, '_':28730, 'end':416, '|':28766, '>':28767, '':13, '<':28789, '|':28766, 'im':321, '_':28730, 'start':2521, '|':28766, '>':28767, 'ass':489, 'istant':11143, '':13 ] [1708357178] sampling: repeat_last_n = 64, repeat_penalty = 1.100, frequency_penalty = 0.000, presence_penalty = 0.000 top_k = 40, tfs_z = 1.000, top_p = 0.950, min_p = 0.050, typical_p = 1.000, temp = 1.000 mirostat = 0, mirostat_lr = 0.100, mirostat_ent = 5.000 [1708357178] sampling order: CFG -> Penalties -> top_k -> tfs_z -> typical_p -> top_p -> min_p -> temperature [1708357178] generate: n_ctx = 512, n_batch = 512, n_predict = 512, n_keep = 0 [1708357178] [1708357178] embd_inp.size(): 10, n_consumed: 0 [1708357178] eval: [ '':1, ' What':1824, ' is':349, ' a':264, ' Large':23292, ' Lang':13550, 'ua':3772, 'ge':490, ' Model':8871, '?':28804 ] [1708357178] n_past = 10 [1708357178] sampled token: 2: '' [1708357178] last: [ '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':0, '':1, ' What':1824, ' is':349, ' a':264, ' Large':23292, ' Lang':13550, 'ua':3772, 'ge':490, ' Model':8871, '?':28804, '':2 ] [1708357178] n_remain: 511 [1708357178] found EOS token [1708357178] [end of text] [1708357178] [1708357178] llama_print_timings: load time = 8832.76 ms [1708357178] llama_print_timings: sample time = 0.36 ms / 1 runs ( 0.36 ms per token, 2808.99 tokens per second) [1708357178] llama_print_timings: prompt eval time = 369.59 ms / 10 tokens ( 36.96 ms per token, 27.06 tokens per second) [1708357178] llama_print_timings: eval time = 0.00 ms / 1 runs ( 0.00 ms per token, inf tokens per second) [1708357178] llama_print_timings: total time = 370.34 ms / 11 tokens [1708357178] Log end