code wiki / (root) / nx_nofloat_qwen_forward_gate.nx

nx_nofloat_qwen_forward_gate.nx

buildroot/runtime/nx_nofloat_qwen_forward_gate.nx

7954 B118 linesdepth 8pulls 17 transitivereach 0 importersview sourcekind gate/prooftopic nofloat
docsdependenciesstructsconstsfunctions

about

nx_nofloat_qwen_forward_gate.nx -- CAPSTONE: a COMPLETE sovereign no-float Qwen2.5-0.5B forward. token ids -> embed (token_embd) -> 24 transformer layers -> final RMSNorm (output_norm) -> LM head (output.weight, streamed over 151936 vocab) -> argmax = predicted next token. ALL building blocks come from the canonical library nx_nofloat_llm.nx (DRY -- no copy-paste debt; one correct accumulate-then-shift matmul). This file is just the wiring + the test. expect_exit: 0 ORIGINAL

dependencies 8 imports · 0 importers

nx_syscalls.nx nx_tier.nx nx_le.nx nx_tensor.nx nx_gguf.nx nx_gguf_load.nx nx_nofloat_llm.nx nx_gate_verdict.nx nx_nofloat_qwen_forward_gate.n

imports: nx_syscalls.nxnx_tier.nxnx_le.nxnx_tensor.nxnx_gguf.nxnx_gguf_load.nxnx_nofloat_llm.nxnx_gate_verdict.nx

imported by: nobody (leaf or entry point)

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

main fw_puts sys_write 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_gguf_find_tensor nx_gguf_tensor_at nx_gguf_name_equals nx_gguf_tensor_at ↻ load_named_q16 nx_gguf_find_tensor ↻ nx_gguf_tensor_at ↻ nx_gguf_tensor_n_values dequant_to_q16 f32_to_q16 nx_le_read_u32 ↻ f16_to_q16 nx_le_read_u16 nx_gguf_dequant_q4_k_q14 nx_gguf_dequant_q4_k nx_le_read_u16 ↻ _gguf_f16_to_q24 nx_le_read_u8 q6k_block _gguf_f16_to_q24 ↻ nx_le_read_u16 ↻

structs

none

consts

none

functions

15func fw_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
16func fw_num(v: i64) -> i64 { let b: *u8=sys_mmap(28); var m: i64=v; if m<0{m=0-m;sys_write(1,"-" as *u8,1)} let t: *u8=sys_mmap(28); var k: i64=0; if m==0{t[0]=48 as u8;k=1} while m>0{t[k]=(48+(m%10)) as u8;m=m/10;k=k+1} var i: i64=0; while i<k{b[i]=t[k-1-i];i=i+1} sys_write(1,b,k); return 0 }
called by 1: main calls 2: sys_mmapsys_write
18func main() -> i64
calls 15: fw_putssys_mmapsys_read_filenx_gguf_parsenx_gguf_find_tensornx_gguf_tensor_at+9