Functions — github.com/andrewkchan/yalm

CodeHub MCPcopy Create free account

hub / github.com/andrewkchan/yalm / functions

Functions105 in github.com/andrewkchan/yalm

⨍ Functions 105◇ Types & classes 21

↓ 21 callersFunctionfill_random

src/test.cpp:128

↓ 18 callersFunctionget_timestamp_ms

↓ 17 callersFunctionerror_usage

src/main.cpp:17

↓ 16 callersFunctioncheck_tensor

src/model.cpp:98

↓ 15 callersFunctionmatmul

src/infer.cpp:48

↓ 13 callersFunctionconv

↓ 13 callersFunctionget_tensor

src/model.cpp:117

↓ 10 callersMethodx

current activations

↓ 10 callersMethodxb

↓ 8 callersMethodforward

src/model.cpp:417

↓ 8 callersMethodhb

↓ 8 callersMethodk

↓ 7 callersFunctionassertArrayEquals

src/test.cpp:29

↓ 7 callersMethodq

↓ 6 callersMethodactive_bytes

src/model.cpp:71

↓ 6 callersMethodcuda

src/model.cpp:215

↓ 5 callersMethodlogits

src/model.h:107

↓ 4 callersMethodatt

src/model.h:100

↓ 4 callersFunctiondtype_size

src/codec.cpp:24

↓ 4 callersFunctiondtype_to_string

src/codec.cpp:9

↓ 4 callersMethodencode

src/tokenizer.cpp:57

↓ 4 callersFunctionhalf_to_float

src/infer.cpp:11

↓ 4 callersMethodv

↓ 4 callersMethodxb2

TODO: do we need xb2?

↓ 3 callersMethodactive_experts

src/model.h:105

↓ 3 callersMethodactive_experts_weights

src/model.h:104

↓ 3 callersFunctionattn

Compute next value in a sequence for a single causal self-attention head.

src/infer.cpp:218

↓ 3 callersFunctionclip

src/infer.cpp:197

↓ 3 callersFunctionfloat_to_half

src/infer.cpp:14

↓ 3 callersMethodfrom_file

src/codec.cpp:97

↓ 3 callersMethodhb2

↓ 3 callersFunctionrmsnorm

src/infer.cpp:136

↓ 3 callersFunctionrope

TODO annotate me

src/infer.cpp:202

↓ 2 callersMethoddecode_one

src/tokenizer.cpp:44

↓ 2 callersMethodencoding_to_debug_string

src/tokenizer.cpp:96

↓ 2 callersFunctionfloat_array_to_half

src/test.cpp:54

↓ 2 callersFunctiongelu

src/infer.cpp:189

↓ 2 callersMethodload

src/debug.cpp:21

↓ 2 callersMethodmoe_weights

mixture of experts

src/model.h:103

↓ 2 callersFunctionpermute_reverse

(w, heads, rotary_dim)

↓ 2 callersFunctionsilu

src/infer.cpp:193

↓ 1 callersFunctionallocateAlignedArray

Helper function to allocate aligned memory

src/test.cpp:237

↓ 1 callersFunctionarrayEquals

src/test.cpp:17

↓ 1 callersMethodblock

src/model.cpp:241

↓ 1 callersMethoddevice

src/model.h:110

↓ 1 callersFunctionffn_cpu

src/infer.cpp:418

↓ 1 callersFunctionfloatEquals

src/test.cpp:13

↓ 1 callersMethodfrom_json

src/codec.cpp:39

↓ 1 callersMethodfrom_yalm

src/model.cpp:16

↓ 1 callersFunctiongpt2_bytes_to_unicode

↓ 1 callersFunctionkernel_bench

src/test.cpp:340

↓ 1 callersFunctionload_tokens

(tokenizer_path, vocab_size)

↓ 1 callersFunctionload_weights

Load all weights from the model files in huggingface format into a dictionary of tensors, normalizing the attention weights, and casting all tens

↓ 1 callersFunctionmatmul_cpu

src/infer.cpp:411

↓ 1 callersMethodmax_err

src/model.cpp:440

↓ 1 callersFunctionmem_bench

src/test.cpp:246

↓ 1 callersFunctionmem_bench2

src/test.cpp:292

↓ 1 callersFunctionmha_cpu

src/infer.cpp:388

↓ 1 callersFunctionmoe_gate

src/infer.cpp:102

↓ 1 callersFunctionrun_completion

src/main.cpp:54

↓ 1 callersFunctionrun_passkey

src/main.cpp:237

↓ 1 callersFunctionrun_perplexity

src/main.cpp:150

↓ 1 callersMethodsample

src/sampler.cpp:40

↓ 1 callersMethodsample_argmax

src/sampler.cpp:27

↓ 1 callersMethodsample_prob

src/sampler.cpp:11

↓ 1 callersFunctionsoftmax

Compute the softmax of an input vector `x` of length `size` and store it in `o`.

src/infer.cpp:172

↓ 1 callersFunctiontest_attn

src/test.cpp:67

↓ 1 callersFunctiontest_cuda_kernels

src/test.cpp:148

↓ 1 callersMethodto_dict

src/model.cpp:128

MethodDebugTensor

src/model.h:279

MethodDebugTensor

src/model.cpp:431

MethodInferenceState

src/model.cpp:284

src/model.cpp:356

src/sampler.cpp:6

MethodTokenizer

src/tokenizer.cpp:3

(self, config, dtype)

Method_block_cpu

src/infer.cpp:257

Method_copy_embedding

src/infer.cpp:454

Method_forward_cpu

src/infer.cpp:477

Functioncopy_debug_tensor

src/infer.cpp:35

Functiondebug_tensors

src/main.cpp:43

Methodkey_cache

src/model.h:184

Functionlayernorm

src/infer.cpp:148

src/test.cpp:407

src/main.cpp:334

Functionmem_bench2_thread

src/test.cpp:285

src/model.h:112

src/model.h:183

Methodrms_att_weight

src/model.h:166

Methodrms_ffn_weight

src/model.h:167

src/debug.cpp:5

Functionsave_debug_tensor

src/infer.cpp:43

src/model.h:113

src/model.h:111

Methodvalue_cache

src/model.h:185

src/model.h:177

src/model.h:179

src/model.h:181

src/model.h:171

next →1–100 of 105, ranked by callers