AetherPrior's picture
Training in progress, epoch 1
b4bec39 verified
{
"_name_or_path": "/checkpoints/postsft-model-v2",
"activation_offloading_max_copy_events": 9,
"activation_offloading_max_num_prefetch": 2,
"architectures": [
"AlexaLlamaForCausalLM"
],
"attention_dropout": 0.0,
"bos_token_id": 1,
"embedding_dropout": 0.0,
"eos_token_id": 2,
"gradient_checkpointing": false,
"hidden_act": "silu",
"hidden_size": 4096,
"initializer_range": 0.02,
"intermediate_size": 16384,
"max_position_embeddings": 2048,
"model_type": "alexallm-v2",
"moe_ffn_hidden_size": null,
"moe_frequency": 1,
"moe_router_activation": "sigmoid",
"moe_shared_experts": 0,
"moe_topk": 1,
"num_attention_heads": 32,
"num_hidden_layers": 32,
"num_key_value_heads": 8,
"num_moe_experts": 1,
"num_query_groups": 8,
"original_max_position_embeddings": 2048,
"pad_token_id": 0,
"position_abf_factor": 1,
"position_interpolation_factor": 1.0,
"residual_dropout": 0.0,
"rms_norm_eps": 1e-06,
"rope_scaling": null,
"rope_state": null,
"rope_theta": 10000.0,
"tie_word_embeddings": true,
"torch_dtype": "bfloat16",
"transformers_version": "4.49.0",
"use_activation_offloading": false,
"use_cache": false,
"use_flash_attention": false,
"use_flash_mlp": false,
"vocab_size": 90000,
"yarn_scale": 1.0
}