Upload model
Browse files- config.json +50 -0
- pytorch_model.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"PerceiverMaskedLanguageModel"
|
| 4 |
+
],
|
| 5 |
+
"model_config": {
|
| 6 |
+
"activation_checkpointing": true,
|
| 7 |
+
"activation_offloading": false,
|
| 8 |
+
"decoder": {
|
| 9 |
+
"cross_attention_residual": false,
|
| 10 |
+
"cross_attention_widening_factor": 1,
|
| 11 |
+
"dropout": 0.1,
|
| 12 |
+
"freeze": false,
|
| 13 |
+
"init_scale": 0.02,
|
| 14 |
+
"max_seq_len": 2048,
|
| 15 |
+
"num_cross_attention_heads": 8,
|
| 16 |
+
"num_cross_attention_qk_channels": 256,
|
| 17 |
+
"num_cross_attention_v_channels": 768,
|
| 18 |
+
"num_output_query_channels": null,
|
| 19 |
+
"vocab_size": 262
|
| 20 |
+
},
|
| 21 |
+
"encoder": {
|
| 22 |
+
"cross_attention_widening_factor": 1,
|
| 23 |
+
"dropout": 0.1,
|
| 24 |
+
"first_cross_attention_layer_shared": false,
|
| 25 |
+
"first_self_attention_block_shared": true,
|
| 26 |
+
"freeze": false,
|
| 27 |
+
"init_scale": 0.02,
|
| 28 |
+
"max_seq_len": 2048,
|
| 29 |
+
"num_cross_attention_heads": 8,
|
| 30 |
+
"num_cross_attention_layers": 1,
|
| 31 |
+
"num_cross_attention_qk_channels": 256,
|
| 32 |
+
"num_cross_attention_v_channels": 1280,
|
| 33 |
+
"num_input_channels": 768,
|
| 34 |
+
"num_self_attention_blocks": 1,
|
| 35 |
+
"num_self_attention_heads": 8,
|
| 36 |
+
"num_self_attention_layers_per_block": 26,
|
| 37 |
+
"num_self_attention_qk_channels": 256,
|
| 38 |
+
"num_self_attention_v_channels": 1280,
|
| 39 |
+
"params": null,
|
| 40 |
+
"self_attention_widening_factor": 1,
|
| 41 |
+
"vocab_size": 262
|
| 42 |
+
},
|
| 43 |
+
"num_latent_channels": 1280,
|
| 44 |
+
"num_latents": 256
|
| 45 |
+
},
|
| 46 |
+
"model_type": "perceiver-io-masked-language-model",
|
| 47 |
+
"tokenizer_class": "PerceiverTokenizer",
|
| 48 |
+
"torch_dtype": "float32",
|
| 49 |
+
"transformers_version": "4.28.0"
|
| 50 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:02233d4f5ac779766f0c4c1fbce57e3bffb4ac9e0abef0a9581016075a8d9428
|
| 3 |
+
size 804586481
|