peagle-gemma4 / config.json.peagle_backup
MeganEFlynn's picture
Upload folder using huggingface_hub
d87b951 verified
Raw
History Blame Contribute Delete
1.73 kB
{
"architectures": [
"PEagleDraftModel"
],
"auto_map": {
"": "config.PEagleSpeculatorConfig"
},
"down_sample_ratio": 0.7,
"down_sample_ratio_min": 0.2,
"draft_vocab_size": 32000,
"dtype": "bfloat16",
"eagle_aux_hidden_state_layer_ids": null,
"embed_requires_grad": true,
"ptd_token_id": 4,
"max_seq_len": 2048,
"norm_before_fc": false,
"norm_before_residual": false,
"para_depths": 4,
"prediction_loss_weight": 1.0,
"speculators_config": {
"algorithm": "peagle",
"default_proposal_method": "greedy",
"proposal_methods": [
{
"accept_tolerance": 0.0,
"proposal_type": "greedy",
"speculative_tokens": 4,
"verifier_accept_k": 1
}
],
"verifier": {
"architectures": [],
"name_or_path": "google/gemma-4-31B-it"
}
},
"speculators_model_type": "peagle",
"speculators_version": "0.5.0.dev100",
"target_hidden_size": null,
"tie_word_embeddings": false,
"transformer_layer_config": {
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 1,
"eos_token_id": 2,
"head_dim": 256,
"hidden_act": "gelu_pytorch_tanh",
"hidden_size": 5376,
"initializer_range": 0.02,
"intermediate_size": 21504,
"max_position_embeddings": 262144,
"mlp_bias": false,
"model_type": "llama",
"num_attention_heads": 32,
"num_hidden_layers": 1,
"num_key_value_heads": 16,
"pad_token_id": null,
"pretraining_tp": 1,
"rms_norm_eps": 1e-06,
"rope_parameters": {
"rope_theta": 10000.0,
"rope_type": "default"
},
"tie_word_embeddings": false,
"use_cache": true,
"vocab_size": 262144
},
"transformers_version": "5.6.2"
}