Upload GPTQ split into three parts to avoid HF upload limit
Browse files- .gitattributes +3 -0
- bloomz-splitaa +3 -0
- bloomz-splitab +3 -0
- bloomz-splitac +3 -0
- config.json +27 -0
- quantize_config.json +10 -0
.gitattributes
CHANGED
@@ -33,3 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
36 |
+
bloomz-splitaa filter=lfs diff=lfs merge=lfs -text
|
37 |
+
bloomz-splitab filter=lfs diff=lfs merge=lfs -text
|
38 |
+
bloomz-splitac filter=lfs diff=lfs merge=lfs -text
|
bloomz-splitaa
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b871915e8ef9ce03dc786d56e9e5071c9408c6557f6ec03d3251d89a882fee56
|
3 |
+
size 48318382080
|
bloomz-splitab
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7366dc7df355fe42a3438f64211e55432a0b8225a6af89114a0a5e9bfc95560e
|
3 |
+
size 48318382080
|
bloomz-splitac
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:476624a94c0ec31b46b08a6364c1147b8ee9e8409dd272ea8c2464e2c3dc21d4
|
3 |
+
size 4145564712
|
config.json
ADDED
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "/home/ubuntu/models/bloomz/",
|
3 |
+
"apply_residual_connection_post_layernorm": false,
|
4 |
+
"architectures": [
|
5 |
+
"BloomForCausalLM"
|
6 |
+
],
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"attention_softmax_in_fp32": true,
|
9 |
+
"bos_token_id": 1,
|
10 |
+
"eos_token_id": 2,
|
11 |
+
"hidden_dropout": 0.0,
|
12 |
+
"hidden_size": 14336,
|
13 |
+
"initializer_range": 0.02,
|
14 |
+
"layer_norm_epsilon": 1e-05,
|
15 |
+
"masked_softmax_fusion": true,
|
16 |
+
"model_type": "bloom",
|
17 |
+
"n_head": 112,
|
18 |
+
"n_layer": 70,
|
19 |
+
"pad_token_id": 3,
|
20 |
+
"pretraining_tp": 4,
|
21 |
+
"seq_length": 2048,
|
22 |
+
"slow_but_exact": false,
|
23 |
+
"torch_dtype": "bfloat16",
|
24 |
+
"transformers_version": "4.28.1",
|
25 |
+
"use_cache": true,
|
26 |
+
"vocab_size": 250880
|
27 |
+
}
|
quantize_config.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bits": 4,
|
3 |
+
"group_size": -1,
|
4 |
+
"damp_percent": 0.1,
|
5 |
+
"desc_act": true,
|
6 |
+
"sym": true,
|
7 |
+
"true_sequential": true,
|
8 |
+
"model_name_or_path": null,
|
9 |
+
"model_file_base_name": null
|
10 |
+
}
|