Skip to main content
Back to top
Ctrl
+
K
gemma
Search
Ctrl
+
K
Gemma
Guides
Sampling
Multi-modal
Finetuning
Tool Use
Sharding
Tokenizer
LoRA (Sampling)
LoRA (Finetuning)
Parameter Efficient Fine Tuning (PEFT)
Checkpoints
Research
Links
GitHub
Issues
Ecosystem
API
gm
gm.ckpts
gm.ckpts.AnchoredPolicyLoader
gm.ckpts.CheckpointPath
gm.ckpts.LoadCheckpoint
gm.ckpts.SkipLoRA
gm.ckpts.load_params
gm.ckpts.save_params
gm.data
gm.data.AddSeq2SeqFields
gm.data.ContrastiveTask
gm.data.DecodeBytes
gm.data.FormatText
gm.data.MapInts
gm.data.Pad
gm.data.Parquet
gm.data.Seq2SeqTask
gm.data.Tokenize
gm.data.make_seq2seq_fields
gm.data.pad
gm.evals
gm.evals.SamplerEvaluator
gm.losses
gm.losses.DpoLoss
gm.losses.NpoLoss
gm.math
gm.math.apply_rope
gm.math.count_consecutive
gm.nn
gm.nn.AnchoredPolicy
gm.nn.AnchoredPolicyOutput
gm.nn.Attention
gm.nn.AttentionType
gm.nn.Block
gm.nn.Cache
gm.nn.Einsum
gm.nn.Embedder
gm.nn.FeedForward
gm.nn.Gemma2_27B
gm.nn.Gemma2_2B
gm.nn.Gemma2_9B
gm.nn.Gemma3_12B
gm.nn.Gemma3_1B
gm.nn.Gemma3_270M
gm.nn.Gemma3_27B
gm.nn.Gemma3_4B
gm.nn.Gemma3n_E2B
gm.nn.Gemma3n_E4B
gm.nn.Gemma4_26B_A4B
gm.nn.Gemma4_31B
gm.nn.Gemma4_E2B
gm.nn.Gemma4_E4B
gm.nn.IntWrapper
gm.nn.LoRA
gm.nn.Output
gm.nn.QuantizationAwareWrapper
gm.nn.RMSNorm
gm.nn.SigLiPFromPatches
gm.nn.Transformer
gm.nn.TransformerLike
gm.nn.config
gm.sharding
gm.testing
gm.testing.DummyGemma
gm.testing.DummyTokenizer
gm.testing.use_hermetic_tokenizer
gm.text
gm.text.ChatSampler
gm.text.Gemma2Tokenizer
gm.text.Gemma3Tokenizer
gm.text.Gemma3nTokenizer
gm.text.Gemma4Sampler
gm.text.Gemma4Tokenizer
gm.text.Greedy
gm.text.RandomSampling
gm.text.Sampler
gm.text.SamplingMethod
gm.text.SpecialTokens
gm.text.Tokenizer
gm.text.ToolSampler
gm.text.TopPSampling
gm.text.TopkSampling
gm.tools
gm.tools.McpToolHandler
gm.tools.ToolHandlerBase
gm.typing
gm.typing.Params
peft
peft.IntDense
peft.IntEinsum
peft.Interceptor
peft.LoRADense
peft.LoRADenseAdapter
peft.LoRADenseGeneral
peft.LoRADenseGeneralAdapter
peft.LoRAEinsum
peft.LoRAEinsumAdapter
peft.ModuleInterceptor
peft.QuantizationMethod
peft.SimulateQuantizedDense
peft.SimulateQuantizedEinsum
peft.fuse_params
peft.get_axis_to_reduce_from_einsum_str
peft.merge_params
peft.quantize
peft.simulate_quantize
peft.split_params
peft.unfuse_params
.md
.pdf
gm.nn.Cache
gm.nn.Cache
#
[[Source]]
Cache
=
dict
[
str
,
_modules
.
LayerCache
]