bayang commited on
Commit
b73c421
1 Parent(s): 2b4accb

Add xl variant

Browse files

contains:
- config-flan-t5-xl.json
- model-flan-t5-xl.gguf

.gitattributes CHANGED
@@ -36,3 +36,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
36
  model.gguf filter=lfs diff=lfs merge=lfs -text
37
  model-flan-t5-large.gguf filter=lfs diff=lfs merge=lfs -text
38
  model-flan-t5-base.gguf filter=lfs diff=lfs merge=lfs -text
 
 
36
  model.gguf filter=lfs diff=lfs merge=lfs -text
37
  model-flan-t5-large.gguf filter=lfs diff=lfs merge=lfs -text
38
  model-flan-t5-base.gguf filter=lfs diff=lfs merge=lfs -text
39
+ model-flan-t5-xl.gguf filter=lfs diff=lfs merge=lfs -text
config-flan-t5-xl.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "T5ForConditionalGeneration"
4
+ ],
5
+ "d_ff": 5120,
6
+ "d_kv": 64,
7
+ "d_model": 2048,
8
+ "decoder_start_token_id": 0,
9
+ "dropout_rate": 0.1,
10
+ "eos_token_id": 1,
11
+ "feed_forward_proj": "gated-gelu",
12
+ "initializer_factor": 1.0,
13
+ "is_encoder_decoder": true,
14
+ "layer_norm_epsilon": 1e-06,
15
+ "model_type": "t5",
16
+ "n_positions": 512,
17
+ "num_decoder_layers": 24,
18
+ "num_heads": 32,
19
+ "num_layers": 24,
20
+ "output_past": true,
21
+ "pad_token_id": 0,
22
+ "relative_attention_max_distance": 128,
23
+ "relative_attention_num_buckets": 32,
24
+ "task_specific_params": {
25
+ "summarization": {
26
+ "early_stopping": true,
27
+ "length_penalty": 2.0,
28
+ "max_length": 200,
29
+ "min_length": 30,
30
+ "no_repeat_ngram_size": 3,
31
+ "num_beams": 4,
32
+ "prefix": "summarize: "
33
+ },
34
+ "translation_en_to_de": {
35
+ "early_stopping": true,
36
+ "max_length": 300,
37
+ "num_beams": 4,
38
+ "prefix": "translate English to German: "
39
+ },
40
+ "translation_en_to_fr": {
41
+ "early_stopping": true,
42
+ "max_length": 300,
43
+ "num_beams": 4,
44
+ "prefix": "translate English to French: "
45
+ },
46
+ "translation_en_to_ro": {
47
+ "early_stopping": true,
48
+ "max_length": 300,
49
+ "num_beams": 4,
50
+ "prefix": "translate English to Romanian: "
51
+ }
52
+ },
53
+ "tie_word_embeddings": false,
54
+ "torch_dtype": "float32",
55
+ "transformers_version": "4.24.0.dev0",
56
+ "use_cache": true,
57
+ "vocab_size": 32128
58
+ }
model-flan-t5-xl.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3ea51bdcd9d9137ef9da0ca5a14ccfb2afd1938d234c089de0787ff980e5bb7c
3
+ size 2338539840