Dataset Viewer
Auto-converted to Parquet Duplicate
identifier
stringlengths
24
117
embedding
listlengths
2.56k
2.56k
tokens
listlengths
4
444
helium/modeling_helium.py:HeliumRMSNorm
[ -0.00009709988080430776, 0.04203189164400101, 0.03231484070420265, 0.053556762635707855, -0.0003689795557875186, 0.040450043976306915, 0.023388713598251343, -0.028925171121954918, 0.008869633078575134, 0.04112797975540161, 0.020789967849850655, 0.006412122864276171, 0.0027258584741503, 0.0...
[ "ModelRMSNorm", "Module", "Parameter", "True", "__init__", "class", "def", "eps", "extra_repr", "f", "float32", "forward", "hidden_size", "hidden_states", "keepdim", "mean", "nn", "ones", "pow", "return", "rsqrt", "self", "shape", "super", "to", "torch", "tuple", ...
helium/modeling_helium.py:HeliumRotaryEmbedding
[ -0.0003748119343072176, 0.04797592759132385, 0.00013228657189756632, -0.009054280817508698, -0.0019108060514554381, 0.0378633551299572, 0.04374275729060173, 0.0032630686182528734, -0.00473291939124465, 0.024105552583932877, -0.0011317851021885872, -0.0009921492310240865, -0.00199899706058204...
[ "False", "ModelRotaryEmbedding", "Module", "None", "ROPE_INIT_FUNCTIONS", "Tensor", "__init__", "and", "arange", "attention_factor", "attention_scaling", "base", "cat", "class", "clone", "compute_default_rope_parameters", "config", "cos", "cpu", "def", "default", "deprecate...
helium/modeling_helium.py:HeliumMLP
[ -0.00024346320424228907, 0.025046052411198616, 0.022045142948627472, 0.02608482912182808, -0.0009449979406781495, 0.060479868203401566, 0.03601091355085373, -0.005799834616482258, -0.00044544751290231943, -0.004155105445533991, 0.028162380680441856, -0.04709119349718094, -0.00169522536452859...
[ "ACT2FN", "Linear", "ModelMLP", "Module", "__init__", "act_fn", "class", "config", "def", "down_proj", "forward", "gate_proj", "hidden_act", "hidden_size", "intermediate_size", "nn", "return", "self", "super", "up_proj", "x" ]
helium/modeling_helium.py:repeat_kv
[-0.0002073117793770507,-0.001958739012479782,-0.003831693669781089,-0.009035934694111347,-0.0004682(...TRUNCATED)
["Model_kv","None","batch","def","expand","head_dim","hidden_states","if","n_rep","num_key_value_hea(...TRUNCATED)
helium/modeling_helium.py:eager_attention_forward
[-0.000020655716070905328,0.020133469253778458,0.01408211700618267,-0.01877615600824356,-0.0,0.03845(...TRUNCATED)
["Model_attention_forward","None","attention_mask","attn_output","attn_weights","contiguous","def","(...TRUNCATED)
helium/modeling_helium.py:rotate_half
[0.000020890642190352082,0.007993859238922596,0.029498467221856117,-0.00021638307953253388,0.0004468(...TRUNCATED)
[ "Model_half", "def", "dim", "flatten", "return", "stack", "torch", "x", "x1", "x2" ]
helium/modeling_helium.py:apply_rotary_pos_emb
[-0.0001693858066573739,0.02724362164735794,0.02361113950610161,-0.0022419230081140995,-0.0007697741(...TRUNCATED)
["Model_rotary_pos_emb","cos","def","dim","k","k_embed","q","q_embed","repeat_interleave","return","(...TRUNCATED)
helium/modeling_helium.py:HeliumAttention
[-0.0001134383783210069,0.03264211490750313,0.0269016046077013,-0.005993767641484737,-0.000481893279(...TRUNCATED)
["ALL_ATTENTION_FUNCTIONS","Linear","ModelAttention","Module","None","Tensor","True","__init__","_at(...TRUNCATED)
helium/modeling_helium.py:HeliumDecoderLayer
[-0.00019977198098786175,0.0459369495511055,0.01272883452475071,0.004129799548536539,-0.000661192229(...TRUNCATED)
["False","GradientCheckpointingLayer","ModelAttention","ModelDecoderLayer","ModelMLP","ModelRMSNorm"(...TRUNCATED)
helium/modeling_helium.py:HeliumPreTrainedModel
[-0.00033686848473735154,0.02805064059793949,0.016459466889500618,0.011127526871860027,-0.0016807202(...TRUNCATED)
["ModelAttention","ModelConfig","ModelDecoderLayer","ModelPreTrainedModel","PreTrainedModel","True",(...TRUNCATED)
End of preview. Expand in Data Studio
README.md exists but content is empty.
Downloads last month
43