Teutonic-I: randomly initialized ~1B Gemma 4 architecture
Browse files- config.json +65 -0
- model.safetensors +3 -0
config.json
ADDED
|
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Gemma4ForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"attention_k_eq_v": false,
|
| 8 |
+
"bos_token_id": 2,
|
| 9 |
+
"dtype": "float32",
|
| 10 |
+
"enable_moe_block": false,
|
| 11 |
+
"eos_token_id": 1,
|
| 12 |
+
"final_logit_softcapping": null,
|
| 13 |
+
"global_head_dim": 512,
|
| 14 |
+
"head_dim": 256,
|
| 15 |
+
"hidden_activation": "gelu_pytorch_tanh",
|
| 16 |
+
"hidden_size": 1536,
|
| 17 |
+
"hidden_size_per_layer_input": 64,
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 6144,
|
| 20 |
+
"layer_types": [
|
| 21 |
+
"sliding_attention",
|
| 22 |
+
"sliding_attention",
|
| 23 |
+
"sliding_attention",
|
| 24 |
+
"sliding_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"sliding_attention",
|
| 27 |
+
"sliding_attention",
|
| 28 |
+
"sliding_attention",
|
| 29 |
+
"sliding_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"sliding_attention",
|
| 32 |
+
"full_attention"
|
| 33 |
+
],
|
| 34 |
+
"max_position_embeddings": 131072,
|
| 35 |
+
"model_type": "gemma4_text",
|
| 36 |
+
"moe_intermediate_size": null,
|
| 37 |
+
"num_attention_heads": 8,
|
| 38 |
+
"num_experts": null,
|
| 39 |
+
"num_global_key_value_heads": null,
|
| 40 |
+
"num_hidden_layers": 12,
|
| 41 |
+
"num_key_value_heads": 1,
|
| 42 |
+
"num_kv_shared_layers": 0,
|
| 43 |
+
"pad_token_id": 0,
|
| 44 |
+
"rms_norm_eps": 1e-06,
|
| 45 |
+
"rope_parameters": {
|
| 46 |
+
"full_attention": {
|
| 47 |
+
"partial_rotary_factor": 0.25,
|
| 48 |
+
"rope_theta": 1000000.0,
|
| 49 |
+
"rope_type": "proportional"
|
| 50 |
+
},
|
| 51 |
+
"sliding_attention": {
|
| 52 |
+
"rope_theta": 10000.0,
|
| 53 |
+
"rope_type": "default"
|
| 54 |
+
}
|
| 55 |
+
},
|
| 56 |
+
"sliding_window": 512,
|
| 57 |
+
"tie_word_embeddings": true,
|
| 58 |
+
"top_k_experts": null,
|
| 59 |
+
"transformers_version": "5.5.3",
|
| 60 |
+
"use_bidirectional_attention": null,
|
| 61 |
+
"use_cache": true,
|
| 62 |
+
"use_double_wide_mlp": true,
|
| 63 |
+
"vocab_size": 262144,
|
| 64 |
+
"vocab_size_per_layer_input": 262144
|
| 65 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c5abdd23d25324d222fc9f5176cacb6d22c0d9dab168b0893421f98ae45931d
|
| 3 |
+
size 4214132368
|