Push model using huggingface_hub.
Browse files- README.md +12 -0
- config.json +52 -0
- model.safetensors +3 -0
README.md
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
pipeline_tag: text-generation
|
| 4 |
+
tags:
|
| 5 |
+
- model_hub_mixin
|
| 6 |
+
- pytorch_model_hub_mixin
|
| 7 |
+
---
|
| 8 |
+
|
| 9 |
+
This model has been pushed to the Hub using the [PytorchModelHubMixin](https://huggingface.co/docs/huggingface_hub/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin) integration:
|
| 10 |
+
- Code: [More Information Needed]
|
| 11 |
+
- Paper: [More Information Needed]
|
| 12 |
+
- Docs: [More Information Needed]
|
config.json
ADDED
|
@@ -0,0 +1,52 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"encoder_config": {
|
| 3 |
+
"att_groups": 8,
|
| 4 |
+
"att_heads": 16,
|
| 5 |
+
"att_query_groups": 8,
|
| 6 |
+
"cross_att_type": "sqa",
|
| 7 |
+
"embed_dim": 512,
|
| 8 |
+
"ff_activation": "silu",
|
| 9 |
+
"ff_dim": 1536,
|
| 10 |
+
"ff_dropout": 0.0,
|
| 11 |
+
"num_layers": 21,
|
| 12 |
+
"rope_base": 100000,
|
| 13 |
+
"self_att_type": "sqa",
|
| 14 |
+
"seq_len": 8192,
|
| 15 |
+
"skip_memory_cross_attention": true,
|
| 16 |
+
"stm_size": 4096,
|
| 17 |
+
"use_attention_output_bias": false,
|
| 18 |
+
"use_flash_attention": true,
|
| 19 |
+
"use_gated": true,
|
| 20 |
+
"use_gated_attention": true,
|
| 21 |
+
"vocab_size": 65536
|
| 22 |
+
},
|
| 23 |
+
"memory_attention_config": {
|
| 24 |
+
"att_groups": 8,
|
| 25 |
+
"att_heads": 16,
|
| 26 |
+
"att_query_groups": 8,
|
| 27 |
+
"att_type": "sqa",
|
| 28 |
+
"embed_dim": 512,
|
| 29 |
+
"interlayer_att_groups": 8,
|
| 30 |
+
"interlayer_att_query_groups": 8,
|
| 31 |
+
"interlayer_att_type": "sqa",
|
| 32 |
+
"norm_type": "classic-rms",
|
| 33 |
+
"num_groups": 3,
|
| 34 |
+
"num_layers": 21,
|
| 35 |
+
"residual_gate_type": "elementwise",
|
| 36 |
+
"residual_per_slot_gate": true,
|
| 37 |
+
"seq_len": 8192,
|
| 38 |
+
"stm_size": 4096,
|
| 39 |
+
"use_flash_attention": false,
|
| 40 |
+
"use_gated_residual": true,
|
| 41 |
+
"use_tanh_residual_gate": false
|
| 42 |
+
},
|
| 43 |
+
"memory_attention_variant": "grouped-self-interlayer",
|
| 44 |
+
"tokenizer": null,
|
| 45 |
+
"tokenizer_config": {
|
| 46 |
+
"answer_token_id": 6,
|
| 47 |
+
"bos_token_id": 2,
|
| 48 |
+
"eos_token_id": 3,
|
| 49 |
+
"pad_token_id": 0,
|
| 50 |
+
"query_token_id": 5
|
| 51 |
+
}
|
| 52 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cdbc46cfc72c45840f23b9751c0d953b960b7277301b0fd75636c9eeb830c06e
|
| 3 |
+
size 6099558592
|