Update tiny models for ModernBertDecoderForCausalLM
Browse files- config.json +52 -0
- generation_config.json +10 -0
- model.safetensors +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +23 -0
config.json
ADDED
|
@@ -0,0 +1,52 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"ModernBertDecoderForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 50281,
|
| 8 |
+
"classifier_activation": "gelu",
|
| 9 |
+
"classifier_bias": false,
|
| 10 |
+
"classifier_dropout": 0.0,
|
| 11 |
+
"cls_token_id": 50281,
|
| 12 |
+
"decoder_bias": true,
|
| 13 |
+
"dtype": "float32",
|
| 14 |
+
"embedding_dropout": 0.0,
|
| 15 |
+
"eos_token_id": 50282,
|
| 16 |
+
"hidden_activation": "gelu",
|
| 17 |
+
"hidden_size": 32,
|
| 18 |
+
"initializer_cutoff_factor": 2.0,
|
| 19 |
+
"initializer_range": 0.02,
|
| 20 |
+
"intermediate_size": 32,
|
| 21 |
+
"is_decoder": true,
|
| 22 |
+
"layer_types": [
|
| 23 |
+
"full_attention",
|
| 24 |
+
"sliding_attention"
|
| 25 |
+
],
|
| 26 |
+
"local_attention": 128,
|
| 27 |
+
"max_position_embeddings": 512,
|
| 28 |
+
"mlp_bias": false,
|
| 29 |
+
"mlp_dropout": 0.0,
|
| 30 |
+
"model_type": "modernbert-decoder",
|
| 31 |
+
"norm_bias": false,
|
| 32 |
+
"norm_eps": 1e-05,
|
| 33 |
+
"num_attention_heads": 2,
|
| 34 |
+
"num_hidden_layers": 2,
|
| 35 |
+
"pad_token_id": 50283,
|
| 36 |
+
"rope_parameters": {
|
| 37 |
+
"full_attention": {
|
| 38 |
+
"rope_theta": 160000.0,
|
| 39 |
+
"rope_type": "default"
|
| 40 |
+
},
|
| 41 |
+
"sliding_attention": {
|
| 42 |
+
"rope_theta": 10000.0,
|
| 43 |
+
"rope_type": "default"
|
| 44 |
+
}
|
| 45 |
+
},
|
| 46 |
+
"sep_token_id": 50282,
|
| 47 |
+
"sliding_window": 64,
|
| 48 |
+
"tie_word_embeddings": false,
|
| 49 |
+
"transformers_version": "5.16.0.dev0",
|
| 50 |
+
"use_cache": true,
|
| 51 |
+
"vocab_size": 50368
|
| 52 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50281,
|
| 4 |
+
"eos_token_id": 50282,
|
| 5 |
+
"output_attentions": false,
|
| 6 |
+
"output_hidden_states": false,
|
| 7 |
+
"pad_token_id": 50283,
|
| 8 |
+
"transformers_version": "5.16.0.dev0",
|
| 9 |
+
"use_cache": true
|
| 10 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0dc7f7e3189e23d03e0611ce147127ae17b38ce171f52e81338fe803f845d03f
|
| 3 |
+
size 13160056
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "[CLS]",
|
| 5 |
+
"clean_up_tokenization_spaces": true,
|
| 6 |
+
"cls_token": "[CLS]",
|
| 7 |
+
"eos_token": "[SEP]",
|
| 8 |
+
"errors": "replace",
|
| 9 |
+
"is_causal": true,
|
| 10 |
+
"is_local": true,
|
| 11 |
+
"local_files_only": false,
|
| 12 |
+
"mask_token": "[MASK]",
|
| 13 |
+
"model_input_names": [
|
| 14 |
+
"input_ids",
|
| 15 |
+
"attention_mask"
|
| 16 |
+
],
|
| 17 |
+
"model_max_length": 512,
|
| 18 |
+
"pad_token": "[PAD]",
|
| 19 |
+
"sep_token": "[SEP]",
|
| 20 |
+
"tokenizer_class": "GPTNeoXTokenizer",
|
| 21 |
+
"trim_offsets": true,
|
| 22 |
+
"unk_token": "[UNK]"
|
| 23 |
+
}
|