upload model weights files
Browse files- config.json +60 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +6 -0
- tokenizer.json +0 -0
- tokenizer_config.json +19 -0
config.json
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_dropout": 0.0,
|
| 3 |
+
"architectures": [
|
| 4 |
+
"FullyShardedDataParallel"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"attn_cfg": {
|
| 8 |
+
"attn_mode": "flash_attn_varkv",
|
| 9 |
+
"ffn_expand_ratio": 1.3333333333333333,
|
| 10 |
+
"no_residual_hop": false,
|
| 11 |
+
"num_heads": 16,
|
| 12 |
+
"num_heads_k": 1,
|
| 13 |
+
"num_heads_v": 16,
|
| 14 |
+
"output_gate": "none",
|
| 15 |
+
"rotary_emb_dim": -1,
|
| 16 |
+
"rotary_emb_scale_base": null,
|
| 17 |
+
"routing_topk": -1,
|
| 18 |
+
"skip_glu": false,
|
| 19 |
+
"skip_ssm": false,
|
| 20 |
+
"use_routing_state": false,
|
| 21 |
+
"window_size": 4096
|
| 22 |
+
},
|
| 23 |
+
"bos_token_id": 126080,
|
| 24 |
+
"d_model": 2048,
|
| 25 |
+
"dropout": 0.0,
|
| 26 |
+
"eos_token_id": 126081,
|
| 27 |
+
"initializer_range": 0.02,
|
| 28 |
+
"jetx_type": "jetoptimus",
|
| 29 |
+
"max_position_embeddings": 4096,
|
| 30 |
+
"mixer_cfg": {
|
| 31 |
+
"expand_ratio": 2.0,
|
| 32 |
+
"input_gate": "sigmoid",
|
| 33 |
+
"input_gate_low_rank": "auto",
|
| 34 |
+
"mem_size": 64,
|
| 35 |
+
"mixer_type": "JetFireV1",
|
| 36 |
+
"mode": "fused_chunk",
|
| 37 |
+
"normalize_epsilon": 1e-05,
|
| 38 |
+
"post_norm": "act_norm",
|
| 39 |
+
"scale": "k_norm",
|
| 40 |
+
"select_gate": "softplus",
|
| 41 |
+
"skip_gate": "sigmoid"
|
| 42 |
+
},
|
| 43 |
+
"model_type": "jetx",
|
| 44 |
+
"n_layer": 24,
|
| 45 |
+
"no_weight_decay_on_bias": true,
|
| 46 |
+
"no_weight_decay_on_embedding": true,
|
| 47 |
+
"no_weight_decay_on_norm": false,
|
| 48 |
+
"norm_epsilon": 1e-05,
|
| 49 |
+
"rescale_prenorm_residual": true,
|
| 50 |
+
"residual_in_fp32": true,
|
| 51 |
+
"tie_word_embeddings": true,
|
| 52 |
+
"torch_dtype": "float32",
|
| 53 |
+
"transformers_version": "4.42.4",
|
| 54 |
+
"use_cache": true,
|
| 55 |
+
"use_fast_path": true,
|
| 56 |
+
"use_fused_cross_entropy": true,
|
| 57 |
+
"use_norm_embedding": false,
|
| 58 |
+
"use_scale_embedding": false,
|
| 59 |
+
"vocab_size": 126464
|
| 60 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8d846d7033fdf8c5f824f1581f631de2d5eec694f04c8127d38d68b0a45a0a18
|
| 3 |
+
size 7464659434
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<|startoftext|>",
|
| 3 |
+
"cls_token": "[CLS]",
|
| 4 |
+
"eos_token": "<|endoftext|>",
|
| 5 |
+
"gmask_token": "[gMASK]"
|
| 6 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"bos_token": "<|startoftext|>",
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"cls_token": "[CLS]",
|
| 7 |
+
"eos_token": "<|endoftext|>",
|
| 8 |
+
"gmask_token": "[gMASK]",
|
| 9 |
+
"merges_file": null,
|
| 10 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 11 |
+
"tokenizer_class": "RodimusTokenizer",
|
| 12 |
+
"vocab_file": null,
|
| 13 |
+
"auto_map": {
|
| 14 |
+
"AutoTokenizer": [
|
| 15 |
+
null,
|
| 16 |
+
"tokenization_rodimus_fast.RodimusTokenizer"
|
| 17 |
+
]
|
| 18 |
+
}
|
| 19 |
+
}
|