danieldk-explosion
commited on
Commit
·
1a52f36
1
Parent(s):
d00224b
Add MPT test
Browse files- config.json +30 -0
- generation_config.json +5 -0
- pytorch_model.bin +3 -0
config.json
ADDED
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"MptForCausalLM"
|
4 |
+
],
|
5 |
+
"attn_config": {
|
6 |
+
"model_type": ""
|
7 |
+
},
|
8 |
+
"d_model": 32,
|
9 |
+
"emb_pdrop": 0.0,
|
10 |
+
"embedding_fraction": 1.0,
|
11 |
+
"expansion_ratio": 4,
|
12 |
+
"init_device": "cpu",
|
13 |
+
"initializer_range": 0.02,
|
14 |
+
"layer_norm_epsilon": 1e-05,
|
15 |
+
"learned_pos_emb": true,
|
16 |
+
"logit_scale": null,
|
17 |
+
"max_seq_len": 2048,
|
18 |
+
"model_type": "mpt",
|
19 |
+
"n_heads": 4,
|
20 |
+
"n_layers": 5,
|
21 |
+
"no_bias": true,
|
22 |
+
"norm_type": "low_precision_layernorm",
|
23 |
+
"num_key_value_heads": 2,
|
24 |
+
"resid_pdrop": 0.0,
|
25 |
+
"torch_dtype": "float32",
|
26 |
+
"transformers_version": "4.32.0.dev0",
|
27 |
+
"use_cache": false,
|
28 |
+
"verbose": 0,
|
29 |
+
"vocab_size": 1024
|
30 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"transformers_version": "4.32.0.dev0",
|
4 |
+
"use_cache": false
|
5 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c2b2337635ab3a4cb734bbca9f6476bc30ea7ede92189cdce3a932ec18ad6e1e
|
3 |
+
size 388177
|