versae's picture
Upload Borealis EDU scorer checkpoint and model card
9cba62d verified
Raw History Blame Contribute Delete
1.43 kB
{
"_sliding_window_pattern": 4,
"architectures": [
"Gemma3TextModel"
],
"attention_bias": false,
"attention_dropout": 0.0,
"attn_logit_softcapping": null,
"bos_token_id": 1,
"cls_token_id": 5,
"dtype": "bfloat16",
"eos_token_id": 2,
"final_logit_softcapping": null,
"head_dim": 64,
"hidden_activation": "silu",
"hidden_size": 768,
"initializer_range": 0.02,
"intermediate_size": 3072,
"layer_types": [
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention"
],
"mask_token_id": 4,
"max_position_embeddings": 8192,
"model_type": "gemma3_text",
"num_attention_heads": 12,
"num_hidden_layers": 12,
"num_key_value_heads": 12,
"pad_token_id": 3,
"prefix_token_ids": [
1
],
"query_pre_attn_scalar": 64,
"rms_norm_eps": 1e-07,
"rope_parameters": {
"full_attention": {
"rope_theta": 100000,
"rope_type": "default"
},
"sliding_attention": {
"rope_theta": 10000,
"rope_type": "default"
}
},
"sliding_window": 1024,
"tie_word_embeddings": true,
"transformers_version": "5.17.0",
"unk_token_id": 0,
"use_bidirectional_attention": true,
"use_cache": false,
"vocab_size": 128000
}