Overview:
layer_pattern
src/schema.yaml
Specifications:
base_model
standard_attn
sigmoid_attn
gqa_attn
retnet
retnet_attn
mamba
ode
titan_attn
engram_attn
sparse_transformer_attn
longformer_attn
bigbird_attn
sparsek_attn
nsa_attn
sparge_attn
fasa_attn
gla_attn
deltanet_attn
gated_deltanet_attn
hgrn2_attn
fox_attn
gated_softmax_attn
falcon1_attn
falcon2_attn
falcon3_attn
falcon1a_attn
falcon2a_attn
falcon3a_attn
cca_attn
ccgqa_attn
gma_attn
ssog_attn
ffn_activation_config
train
deploy
infer
web-server
expansion_rate
similarity
search
cluster
encode
optimizer_class
API Reference:
build_parser()
main()
StreamingMLMDataset
SBERTInference
SimilarityResult
ModelDeployer
FrankensteinInference
interactive_mode()
ActivationQuantizer
BitNetQuantizer
bake_bitnet_weights()
estimate_model_size()
load_quantized_checkpoint()
save_quantized_checkpoint()
SpanishSPMTokenizer
is_mps_available()
resolve_torch_device()
GPUTelemetryError
GPUTempCheckResult
GPUTempSupervisor
GPUTemperatureGuard
read_temperature_from_nvidia_smi()
StorageManager
Bibliography:
Technical Reports: