code wiki / (root) / nx_f32_llm_chat_test.nx

nx_f32_llm_chat_test.nx

buildroot/runtime/nx_f32_llm_chat_test.nx

5757 B128 linesdepth 13pulls 59 transitivereach 0 importersview sourcekind gate/prooftopic f32
docsdependenciesstructsconstsfunctions

about

nx_f32_llm_chat_test.nx -- SOVEREIGN CHAT HARNESS: the decisive coherence test. Qwen2.5-Instruct is trained for the chat template; a raw prompt is out-of-distribution and drifts. This builds the REAL template with special-token ids (<|im_start|>=151644, <|im_end|>=151645) -> "<|im_start|>user\n{msg}<|im_end|>\n <|im_start|>assistant\n" -> prefills token-by-token (memory-safe) -> greedy-generates -> decodes (byte-level unmap). If the engine is correct, this produces a coherent English answer. Proves engine-correct vs forward-bug. expect_exit: 0 license_tier: ORIGINAL module: nishi-core.llm.chat

dependencies 18 imports · 0 importers

nx_syscalls.nx nx_tier.nx nx_bpe.nx nx_gguf.nx nx_gguf_load.nx nx_gguf_meta.nx nx_f32.nx nx_f32_kv_cache.nx nx_f32_lazy_weight.nx nx_f32_llama_block.nx nx_f32_llm_chat_test.nx

diagram shows first 10 each side; +8 more imports, +0 more importers in the complete lists below.

imports: nx_syscalls.nxnx_tier.nxnx_bpe.nxnx_gguf.nxnx_gguf_load.nxnx_gguf_meta.nxnx_f32.nxnx_f32_kv_cache.nxnx_f32_lazy_weight.nxnx_f32_llama_block.nxnx_f32_llama_block_v4.nxnx_f32_llama_stack_v4.nxnx_f32_llama_layer_lazy_load.nxnx_f32_llm.nxnx_f32_llm_v4.nxnx_f32_llm_read_dims.nxnx_f32_bpe_load.nxnx_f32_llm_special_tokens.nx

imported by: nobody (leaf or entry point)

call flow from main pre-order; caps 40 nodes / depth 6 declared; ↻ = already shown

main sys_mmap sys_read_file sys_openat_rd sys_lseek sys_mmap ↻ sys_read sys_close nx_gguf_parse nx_le_read_u32 nx_le_read_u64 nx_le_read_u32 ↻ _gguf_skip_value nx_le_read_u64 ↻ nx_le_read_u32 ↻ _gguf_skip_value ↻ sys_mmap ↻ nx_f32_llama_model_alloc sys_mmap ↻ nx_f32_llm_read_dims_from_ sys_mmap ↻ nx_gguf_meta_find nx_le_read_u64 ↻ nx_le_read_u32 ↻ _gmeta_bytes_equal nx_gguf_skip_value _gguf_skip_value ↻ nx_gguf_meta_read_string_l nx_le_read_u64 ↻ nx_gguf_meta_read_string_p _fld_read_arch_u32 sys_mmap ↻ _fld_concat_key nx_gguf_meta_find ↻ nx_gguf_meta_read_u32 nx_le_read_u32 ↻ _fld_read_vocab_size_from_ sys_mmap ↻ nx_gguf_find_tensor nx_gguf_tensor_at

structs

none

consts

none

functions

26func pr_puts(s: *u8) -> i64 { var n: i64=0; while s[n]!=(0 as u8){n=n+1} sys_write(1,s,n); return 0 }
called by 1: main calls 1: sys_write
27func pr_num(v: i64) -> i64
called by 1: main calls 2: sys_mmapsys_write
32func ch_enc(vocab: *NxBpeVocab, text: *u8, tlen: nx_int, toks: *i64, off: nx_int) -> nx_int
called by 1: main calls 2: sys_mmapnx_bpe_encode_bytelevel
39func ch_argmax(logits: *i64, vocab: nx_int) -> nx_int
called by 1: main calls 1: nx_f32_gt
50func main() -> i64
calls 17: sys_mmapsys_read_filenx_gguf_parsenx_f32_llama_model_allocnx_f32_llm_read_dims_from_ggufnx_f32_llm_load_weights_v4_from_gguf+11