Hugging Face's logo

StonyBrookNLP
/
teabreac-t5-3b

Text2Text Generation
Transformers
PyTorch
t5
question-answering, multi-step-reasoning, multi-hop-reasoning
text-generation-inference
Inference Endpoints
Model card Files Community
Harsh Trivedi commited on
Commit
6a3527f
·
1 Parent(s): 274dfc3

update.

Browse files
Files changed (9) hide show
  1. README.md +54 -0
  2. added_tokens.json +68 -0
  3. config.json +60 -0
  4. pytorch_model-00001-of-00002.bin +3 -0
  5. pytorch_model-00002-of-00002.bin +3 -0
  6. pytorch_model.bin.index.json +519 -0
  7. special_tokens_map.json +17 -0
  8. spiece.model +3 -0
  9. tokenizer_config.json +114 -0
README.md ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - question-answering, multi-step-reasoning, multi-hop-reasoning
4
+ thumbnail: https://raw.githubusercontent.com/StonyBrookNLP/teabreac/main/teabreac_icon.png
5
+ license: cc-by-4.0
6
+ ---
7
+
8
+ # What's this?
9
+
10
+ This is one of the models reported in the paper: ["Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts".](https://arxiv.org/abs/2205.12496).
11
+
12
+ This paper proposes a procedure to synthetically generate a QA dataset, TeaBReaC, for pretraining language models for robust multi-step reasoning. Pretraining plain LMs like Bart, T5 and numerate LMs like NT5, PReasM, POET on TeaBReaC leads to improvemed downstream performance on several multi-step QA datasets. Please checkout out the paper for the details.
13
+
14
+ We release the following models:
15
+
16
+ - **A:** Base Models finetuned on target datasets: `{base_model}-{target_dataset}`
17
+ - **B:** Base models pretrained on TeaBReaC: `teabreac-{base_model}`
18
+ - **C:** Base models pretrained on TeaBReaC and then finetuned on target datasets: `teabreac-{base_model}-{target_dataset}`
19
+
20
+ The `base_model` above can be from: `bart-large`, `t5-large`, `t5-3b`, `nt5-small`, `preasm-large`.
21
+ The `target_dataset` above can be from: `drop`, `tatqa`, `iirc-gold`, `iirc-retrieved`, `numglue`.
22
+
23
+ The **A** models are only released for completeness / reproducibility. In your end application you probably just want to use either **B** or **C**.
24
+
25
+ # How to use it?
26
+
27
+ Please checkout the details in our [github repository](https://github.com/stonybrooknlp/teabreac), but in a nutshell:
28
+
29
+ ```python
30
+ # NOTE: This model is only pretrained on TeaBReaC, and not on any real QA dataset.
31
+ from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
32
+ from digit_tokenization import enable_digit_tokenization # digit_tokenization.py from https://github.com/stonybrooknlp/teabreac
33
+
34
+ model_name = "teabreac-t5-3b"
35
+ tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False) # Fast doesn't work with digit tokenization
36
+ model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
37
+ enable_digit_tokenization(tokenizer)
38
+ input_texts = [
39
+ "answer_me: Who scored the first touchdown of the game?" +
40
+ "context: ... Oakland would get the early lead in the first quarter as quarterback JaMarcus Russell completed a 20-yard touchdown pass to rookie wide receiver Chaz Schilens..."
41
+ # Note: some models have slightly different qn/ctxt format. See the github repo.
42
+ ]
43
+ input_ids = tokenizer(
44
+ input_texts, return_tensors="pt",
45
+ truncation=True, max_length=800,
46
+ add_special_tokens=True, padding=True,
47
+ )
48
+ generated_ids = model.generate(input_ids, min_length=1, max_length=50)
49
+ generated_predictions = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
50
+ generated_predictions = [
51
+ tokenizer.fix_decoded_text(generated_prediction) for generated_prediction in generated_predictions
52
+ ]
53
+ # => ["Chaz Schilens"]
54
+ ```
added_tokens.json ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "<ss>": 32146,
3
+ "[CHPD]": 32150,
4
+ "[CND]": 32147,
5
+ "[COLD]": 32140,
6
+ "[CTANS]": 32151,
7
+ "[CTCNINSINA]": 32155,
8
+ "[CTCNINTA]": 32154,
9
+ "[CTOUINSINA]": 32153,
10
+ "[CTOUINTA]": 32152,
11
+ "[DAY]": 32163,
12
+ "[DFB]": 32142,
13
+ "[DFD]": 32144,
14
+ "[DFE]": 32143,
15
+ "[MDFD]": 32145,
16
+ "[MONTH]": 32164,
17
+ "[ROWD]": 32141,
18
+ "[SCBN]": 32160,
19
+ "[SCBP]": 32159,
20
+ "[SCD]": 32158,
21
+ "[SCMN]": 32162,
22
+ "[SCMP]": 32161,
23
+ "[SIND]": 32149,
24
+ "[SNE]": 32157,
25
+ "[SNS]": 32156,
26
+ "[SOMD]": 32148,
27
+ "[YEAR]": 32165,
28
+ "are_items_different": 32103,
29
+ "are_items_same": 32102,
30
+ "arg_bool": 32131,
31
+ "arg_intersection": 32132,
32
+ "arg_maximum_date": 32116,
33
+ "arg_maximum_number": 32105,
34
+ "arg_minimum_date": 32115,
35
+ "arg_minimum_number": 32104,
36
+ "arithmetic_division": 32111,
37
+ "arithmetic_mean_single": 32108,
38
+ "arithmetic_multiplication": 32113,
39
+ "arithmetic_subtraction": 32112,
40
+ "arithmetic_sum_multiple": 32110,
41
+ "arithmetic_sum_single": 32109,
42
+ "boolean": 32101,
43
+ "compare_dates": 32120,
44
+ "compare_numbers": 32114,
45
+ "date_subtraction": 32119,
46
+ "filter_a_where_b_is_compared_to": 32128,
47
+ "filter_a_where_b_is_compared_to_date": 32124,
48
+ "filter_a_where_b_is_given_value": 32127,
49
+ "filter_a_where_b_is_max": 32125,
50
+ "filter_a_where_b_is_max_date": 32122,
51
+ "filter_a_where_b_is_min": 32126,
52
+ "filter_a_where_b_is_min_date": 32123,
53
+ "grouped_count": 32134,
54
+ "grouped_mean": 32135,
55
+ "grouped_sum": 32136,
56
+ "intersection": 32133,
57
+ "kth_highest": 32137,
58
+ "kth_lowest": 32138,
59
+ "list_subtraction": 32121,
60
+ "logical_and": 32130,
61
+ "logical_or": 32129,
62
+ "maximum_date": 32118,
63
+ "maximum_number": 32107,
64
+ "minimum_date": 32117,
65
+ "minimum_number": 32106,
66
+ "null": 32139,
67
+ "select": 32100
68
+ }
config.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "t5-3b",
3
+ "architectures": [
4
+ "T5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 16384,
7
+ "d_kv": 128,
8
+ "d_model": 1024,
9
+ "decoder_start_token_id": 0,
10
+ "dense_act_fn": "relu",
11
+ "dropout_rate": 0.1,
12
+ "eos_token_id": 1,
13
+ "feed_forward_proj": "relu",
14
+ "initializer_factor": 1.0,
15
+ "is_encoder_decoder": true,
16
+ "is_gated_act": false,
17
+ "layer_norm_epsilon": 1e-06,
18
+ "model_type": "t5",
19
+ "n_positions": 512,
20
+ "num_decoder_layers": 24,
21
+ "num_heads": 32,
22
+ "num_layers": 24,
23
+ "output_past": true,
24
+ "pad_token_id": 0,
25
+ "relative_attention_max_distance": 128,
26
+ "relative_attention_num_buckets": 32,
27
+ "task_specific_params": {
28
+ "summarization": {
29
+ "early_stopping": true,
30
+ "length_penalty": 2.0,
31
+ "max_length": 200,
32
+ "min_length": 30,
33
+ "no_repeat_ngram_size": 3,
34
+ "num_beams": 4,
35
+ "prefix": "summarize: "
36
+ },
37
+ "translation_en_to_de": {
38
+ "early_stopping": true,
39
+ "max_length": 300,
40
+ "num_beams": 4,
41
+ "prefix": "translate English to German: "
42
+ },
43
+ "translation_en_to_fr": {
44
+ "early_stopping": true,
45
+ "max_length": 300,
46
+ "num_beams": 4,
47
+ "prefix": "translate English to French: "
48
+ },
49
+ "translation_en_to_ro": {
50
+ "early_stopping": true,
51
+ "max_length": 300,
52
+ "num_beams": 4,
53
+ "prefix": "translate English to Romanian: "
54
+ }
55
+ },
56
+ "torch_dtype": "float32",
57
+ "transformers_version": "4.24.0",
58
+ "use_cache": true,
59
+ "vocab_size": 32166
60
+ }
pytorch_model-00001-of-00002.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf89aa25733e72456f93bce5cafd915610cc3300f840386cb00f2a8451d55743
3
+ size 9728887645
pytorch_model-00002-of-00002.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:faf731cc4ba52bafd17008b4c0f355f85d7b0d001c10b1f084d8a3d85c7eda5d
3
+ size 1809581167
pytorch_model.bin.index.json ADDED
@@ -0,0 +1,519 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 11801804800
4
+ },
5
+ "weight_map": {
6
+ "decoder.block.0.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
7
+ "decoder.block.0.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
8
+ "decoder.block.0.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
9
+ "decoder.block.0.layer.0.SelfAttention.relative_attention_bias.weight": "pytorch_model-00001-of-00002.bin",
10
+ "decoder.block.0.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
11
+ "decoder.block.0.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
12
+ "decoder.block.0.layer.1.EncDecAttention.k.weight": "pytorch_model-00001-of-00002.bin",
13
+ "decoder.block.0.layer.1.EncDecAttention.o.weight": "pytorch_model-00001-of-00002.bin",
14
+ "decoder.block.0.layer.1.EncDecAttention.q.weight": "pytorch_model-00001-of-00002.bin",
15
+ "decoder.block.0.layer.1.EncDecAttention.v.weight": "pytorch_model-00001-of-00002.bin",
16
+ "decoder.block.0.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
17
+ "decoder.block.0.layer.2.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
18
+ "decoder.block.0.layer.2.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
19
+ "decoder.block.0.layer.2.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
20
+ "decoder.block.1.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
21
+ "decoder.block.1.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
22
+ "decoder.block.1.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
23
+ "decoder.block.1.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
24
+ "decoder.block.1.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
25
+ "decoder.block.1.layer.1.EncDecAttention.k.weight": "pytorch_model-00001-of-00002.bin",
26
+ "decoder.block.1.layer.1.EncDecAttention.o.weight": "pytorch_model-00001-of-00002.bin",
27
+ "decoder.block.1.layer.1.EncDecAttention.q.weight": "pytorch_model-00001-of-00002.bin",
28
+ "decoder.block.1.layer.1.EncDecAttention.v.weight": "pytorch_model-00001-of-00002.bin",
29
+ "decoder.block.1.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
30
+ "decoder.block.1.layer.2.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
31
+ "decoder.block.1.layer.2.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
32
+ "decoder.block.1.layer.2.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
33
+ "decoder.block.10.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
34
+ "decoder.block.10.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
35
+ "decoder.block.10.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
36
+ "decoder.block.10.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
37
+ "decoder.block.10.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
38
+ "decoder.block.10.layer.1.EncDecAttention.k.weight": "pytorch_model-00001-of-00002.bin",
39
+ "decoder.block.10.layer.1.EncDecAttention.o.weight": "pytorch_model-00001-of-00002.bin",
40
+ "decoder.block.10.layer.1.EncDecAttention.q.weight": "pytorch_model-00001-of-00002.bin",
41
+ "decoder.block.10.layer.1.EncDecAttention.v.weight": "pytorch_model-00001-of-00002.bin",
42
+ "decoder.block.10.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
43
+ "decoder.block.10.layer.2.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
44
+ "decoder.block.10.layer.2.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
45
+ "decoder.block.10.layer.2.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
46
+ "decoder.block.11.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
47
+ "decoder.block.11.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
48
+ "decoder.block.11.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
49
+ "decoder.block.11.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
50
+ "decoder.block.11.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
51
+ "decoder.block.11.layer.1.EncDecAttention.k.weight": "pytorch_model-00001-of-00002.bin",
52
+ "decoder.block.11.layer.1.EncDecAttention.o.weight": "pytorch_model-00001-of-00002.bin",
53
+ "decoder.block.11.layer.1.EncDecAttention.q.weight": "pytorch_model-00001-of-00002.bin",
54
+ "decoder.block.11.layer.1.EncDecAttention.v.weight": "pytorch_model-00001-of-00002.bin",
55
+ "decoder.block.11.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
56
+ "decoder.block.11.layer.2.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
57
+ "decoder.block.11.layer.2.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
58
+ "decoder.block.11.layer.2.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
59
+ "decoder.block.12.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
60
+ "decoder.block.12.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
61
+ "decoder.block.12.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
62
+ "decoder.block.12.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
63
+ "decoder.block.12.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
64
+ "decoder.block.12.layer.1.EncDecAttention.k.weight": "pytorch_model-00001-of-00002.bin",
65
+ "decoder.block.12.layer.1.EncDecAttention.o.weight": "pytorch_model-00001-of-00002.bin",
66
+ "decoder.block.12.layer.1.EncDecAttention.q.weight": "pytorch_model-00001-of-00002.bin",
67
+ "decoder.block.12.layer.1.EncDecAttention.v.weight": "pytorch_model-00001-of-00002.bin",
68
+ "decoder.block.12.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
69
+ "decoder.block.12.layer.2.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
70
+ "decoder.block.12.layer.2.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
71
+ "decoder.block.12.layer.2.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
72
+ "decoder.block.13.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
73
+ "decoder.block.13.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
74
+ "decoder.block.13.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
75
+ "decoder.block.13.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
76
+ "decoder.block.13.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
77
+ "decoder.block.13.layer.1.EncDecAttention.k.weight": "pytorch_model-00001-of-00002.bin",
78
+ "decoder.block.13.layer.1.EncDecAttention.o.weight": "pytorch_model-00001-of-00002.bin",
79
+ "decoder.block.13.layer.1.EncDecAttention.q.weight": "pytorch_model-00001-of-00002.bin",
80
+ "decoder.block.13.layer.1.EncDecAttention.v.weight": "pytorch_model-00001-of-00002.bin",
81
+ "decoder.block.13.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
82
+ "decoder.block.13.layer.2.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
83
+ "decoder.block.13.layer.2.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
84
+ "decoder.block.13.layer.2.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
85
+ "decoder.block.14.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
86
+ "decoder.block.14.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
87
+ "decoder.block.14.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
88
+ "decoder.block.14.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
89
+ "decoder.block.14.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
90
+ "decoder.block.14.layer.1.EncDecAttention.k.weight": "pytorch_model-00001-of-00002.bin",
91
+ "decoder.block.14.layer.1.EncDecAttention.o.weight": "pytorch_model-00001-of-00002.bin",
92
+ "decoder.block.14.layer.1.EncDecAttention.q.weight": "pytorch_model-00001-of-00002.bin",
93
+ "decoder.block.14.layer.1.EncDecAttention.v.weight": "pytorch_model-00001-of-00002.bin",
94
+ "decoder.block.14.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
95
+ "decoder.block.14.layer.2.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
96
+ "decoder.block.14.layer.2.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
97
+ "decoder.block.14.layer.2.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
98
+ "decoder.block.15.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
99
+ "decoder.block.15.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
100
+ "decoder.block.15.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
101
+ "decoder.block.15.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
102
+ "decoder.block.15.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
103
+ "decoder.block.15.layer.1.EncDecAttention.k.weight": "pytorch_model-00001-of-00002.bin",
104
+ "decoder.block.15.layer.1.EncDecAttention.o.weight": "pytorch_model-00001-of-00002.bin",
105
+ "decoder.block.15.layer.1.EncDecAttention.q.weight": "pytorch_model-00001-of-00002.bin",
106
+ "decoder.block.15.layer.1.EncDecAttention.v.weight": "pytorch_model-00001-of-00002.bin",
107
+ "decoder.block.15.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
108
+ "decoder.block.15.layer.2.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
109
+ "decoder.block.15.layer.2.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
110
+ "decoder.block.15.layer.2.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
111
+ "decoder.block.16.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
112
+ "decoder.block.16.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
113
+ "decoder.block.16.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
114
+ "decoder.block.16.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
115
+ "decoder.block.16.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
116
+ "decoder.block.16.layer.1.EncDecAttention.k.weight": "pytorch_model-00001-of-00002.bin",
117
+ "decoder.block.16.layer.1.EncDecAttention.o.weight": "pytorch_model-00001-of-00002.bin",
118
+ "decoder.block.16.layer.1.EncDecAttention.q.weight": "pytorch_model-00001-of-00002.bin",
119
+ "decoder.block.16.layer.1.EncDecAttention.v.weight": "pytorch_model-00001-of-00002.bin",
120
+ "decoder.block.16.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
121
+ "decoder.block.16.layer.2.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
122
+ "decoder.block.16.layer.2.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
123
+ "decoder.block.16.layer.2.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
124
+ "decoder.block.17.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
125
+ "decoder.block.17.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
126
+ "decoder.block.17.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
127
+ "decoder.block.17.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
128
+ "decoder.block.17.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
129
+ "decoder.block.17.layer.1.EncDecAttention.k.weight": "pytorch_model-00001-of-00002.bin",
130
+ "decoder.block.17.layer.1.EncDecAttention.o.weight": "pytorch_model-00001-of-00002.bin",
131
+ "decoder.block.17.layer.1.EncDecAttention.q.weight": "pytorch_model-00001-of-00002.bin",
132
+ "decoder.block.17.layer.1.EncDecAttention.v.weight": "pytorch_model-00001-of-00002.bin",
133
+ "decoder.block.17.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
134
+ "decoder.block.17.layer.2.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
135
+ "decoder.block.17.layer.2.DenseReluDense.wo.weight": "pytorch_model-00002-of-00002.bin",
136
+ "decoder.block.17.layer.2.layer_norm.weight": "pytorch_model-00002-of-00002.bin",
137
+ "decoder.block.18.layer.0.SelfAttention.k.weight": "pytorch_model-00002-of-00002.bin",
138
+ "decoder.block.18.layer.0.SelfAttention.o.weight": "pytorch_model-00002-of-00002.bin",
139
+ "decoder.block.18.layer.0.SelfAttention.q.weight": "pytorch_model-00002-of-00002.bin",
140
+ "decoder.block.18.layer.0.SelfAttention.v.weight": "pytorch_model-00002-of-00002.bin",
141
+ "decoder.block.18.layer.0.layer_norm.weight": "pytorch_model-00002-of-00002.bin",
142
+ "decoder.block.18.layer.1.EncDecAttention.k.weight": "pytorch_model-00002-of-00002.bin",
143
+ "decoder.block.18.layer.1.EncDecAttention.o.weight": "pytorch_model-00002-of-00002.bin",
144
+ "decoder.block.18.layer.1.EncDecAttention.q.weight": "pytorch_model-00002-of-00002.bin",
145
+ "decoder.block.18.layer.1.EncDecAttention.v.weight": "pytorch_model-00002-of-00002.bin",
146
+ "decoder.block.18.layer.1.layer_norm.weight": "pytorch_model-00002-of-00002.bin",
147
+ "decoder.block.18.layer.2.DenseReluDense.wi.weight": "pytorch_model-00002-of-00002.bin",
148
+ "decoder.block.18.layer.2.DenseReluDense.wo.weight": "pytorch_model-00002-of-00002.bin",
149
+ "decoder.block.18.layer.2.layer_norm.weight": "pytorch_model-00002-of-00002.bin",
150
+ "decoder.block.19.layer.0.SelfAttention.k.weight": "pytorch_model-00002-of-00002.bin",
151
+ "decoder.block.19.layer.0.SelfAttention.o.weight": "pytorch_model-00002-of-00002.bin",
152
+ "decoder.block.19.layer.0.SelfAttention.q.weight": "pytorch_model-00002-of-00002.bin",
153
+ "decoder.block.19.layer.0.SelfAttention.v.weight": "pytorch_model-00002-of-00002.bin",
154
+ "decoder.block.19.layer.0.layer_norm.weight": "pytorch_model-00002-of-00002.bin",
155
+ "decoder.block.19.layer.1.EncDecAttention.k.weight": "pytorch_model-00002-of-00002.bin",
156
+ "decoder.block.19.layer.1.EncDecAttention.o.weight": "pytorch_model-00002-of-00002.bin",
157
+ "decoder.block.19.layer.1.EncDecAttention.q.weight": "pytorch_model-00002-of-00002.bin",
158
+ "decoder.block.19.layer.1.EncDecAttention.v.weight": "pytorch_model-00002-of-00002.bin",
159
+ "decoder.block.19.layer.1.layer_norm.weight": "pytorch_model-00002-of-00002.bin",
160
+ "decoder.block.19.layer.2.DenseReluDense.wi.weight": "pytorch_model-00002-of-00002.bin",
161
+ "decoder.block.19.layer.2.DenseReluDense.wo.weight": "pytorch_model-00002-of-00002.bin",
162
+ "decoder.block.19.layer.2.layer_norm.weight": "pytorch_model-00002-of-00002.bin",
163
+ "decoder.block.2.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
164
+ "decoder.block.2.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
165
+ "decoder.block.2.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
166
+ "decoder.block.2.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
167
+ "decoder.block.2.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
168
+ "decoder.block.2.layer.1.EncDecAttention.k.weight": "pytorch_model-00001-of-00002.bin",
169
+ "decoder.block.2.layer.1.EncDecAttention.o.weight": "pytorch_model-00001-of-00002.bin",
170
+ "decoder.block.2.layer.1.EncDecAttention.q.weight": "pytorch_model-00001-of-00002.bin",
171
+ "decoder.block.2.layer.1.EncDecAttention.v.weight": "pytorch_model-00001-of-00002.bin",
172
+ "decoder.block.2.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
173
+ "decoder.block.2.layer.2.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
174
+ "decoder.block.2.layer.2.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
175
+ "decoder.block.2.layer.2.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
176
+ "decoder.block.20.layer.0.SelfAttention.k.weight": "pytorch_model-00002-of-00002.bin",
177
+ "decoder.block.20.layer.0.SelfAttention.o.weight": "pytorch_model-00002-of-00002.bin",
178
+ "decoder.block.20.layer.0.SelfAttention.q.weight": "pytorch_model-00002-of-00002.bin",
179
+ "decoder.block.20.layer.0.SelfAttention.v.weight": "pytorch_model-00002-of-00002.bin",
180
+ "decoder.block.20.layer.0.layer_norm.weight": "pytorch_model-00002-of-00002.bin",
181
+ "decoder.block.20.layer.1.EncDecAttention.k.weight": "pytorch_model-00002-of-00002.bin",
182
+ "decoder.block.20.layer.1.EncDecAttention.o.weight": "pytorch_model-00002-of-00002.bin",
183
+ "decoder.block.20.layer.1.EncDecAttention.q.weight": "pytorch_model-00002-of-00002.bin",
184
+ "decoder.block.20.layer.1.EncDecAttention.v.weight": "pytorch_model-00002-of-00002.bin",
185
+ "decoder.block.20.layer.1.layer_norm.weight": "pytorch_model-00002-of-00002.bin",
186
+ "decoder.block.20.layer.2.DenseReluDense.wi.weight": "pytorch_model-00002-of-00002.bin",
187
+ "decoder.block.20.layer.2.DenseReluDense.wo.weight": "pytorch_model-00002-of-00002.bin",
188
+ "decoder.block.20.layer.2.layer_norm.weight": "pytorch_model-00002-of-00002.bin",
189
+ "decoder.block.21.layer.0.SelfAttention.k.weight": "pytorch_model-00002-of-00002.bin",
190
+ "decoder.block.21.layer.0.SelfAttention.o.weight": "pytorch_model-00002-of-00002.bin",
191
+ "decoder.block.21.layer.0.SelfAttention.q.weight": "pytorch_model-00002-of-00002.bin",
192
+ "decoder.block.21.layer.0.SelfAttention.v.weight": "pytorch_model-00002-of-00002.bin",
193
+ "decoder.block.21.layer.0.layer_norm.weight": "pytorch_model-00002-of-00002.bin",
194
+ "decoder.block.21.layer.1.EncDecAttention.k.weight": "pytorch_model-00002-of-00002.bin",
195
+ "decoder.block.21.layer.1.EncDecAttention.o.weight": "pytorch_model-00002-of-00002.bin",
196
+ "decoder.block.21.layer.1.EncDecAttention.q.weight": "pytorch_model-00002-of-00002.bin",
197
+ "decoder.block.21.layer.1.EncDecAttention.v.weight": "pytorch_model-00002-of-00002.bin",
198
+ "decoder.block.21.layer.1.layer_norm.weight": "pytorch_model-00002-of-00002.bin",
199
+ "decoder.block.21.layer.2.DenseReluDense.wi.weight": "pytorch_model-00002-of-00002.bin",
200
+ "decoder.block.21.layer.2.DenseReluDense.wo.weight": "pytorch_model-00002-of-00002.bin",
201
+ "decoder.block.21.layer.2.layer_norm.weight": "pytorch_model-00002-of-00002.bin",
202
+ "decoder.block.22.layer.0.SelfAttention.k.weight": "pytorch_model-00002-of-00002.bin",
203
+ "decoder.block.22.layer.0.SelfAttention.o.weight": "pytorch_model-00002-of-00002.bin",
204
+ "decoder.block.22.layer.0.SelfAttention.q.weight": "pytorch_model-00002-of-00002.bin",
205
+ "decoder.block.22.layer.0.SelfAttention.v.weight": "pytorch_model-00002-of-00002.bin",
206
+ "decoder.block.22.layer.0.layer_norm.weight": "pytorch_model-00002-of-00002.bin",
207
+ "decoder.block.22.layer.1.EncDecAttention.k.weight": "pytorch_model-00002-of-00002.bin",
208
+ "decoder.block.22.layer.1.EncDecAttention.o.weight": "pytorch_model-00002-of-00002.bin",
209
+ "decoder.block.22.layer.1.EncDecAttention.q.weight": "pytorch_model-00002-of-00002.bin",
210
+ "decoder.block.22.layer.1.EncDecAttention.v.weight": "pytorch_model-00002-of-00002.bin",
211
+ "decoder.block.22.layer.1.layer_norm.weight": "pytorch_model-00002-of-00002.bin",
212
+ "decoder.block.22.layer.2.DenseReluDense.wi.weight": "pytorch_model-00002-of-00002.bin",
213
+ "decoder.block.22.layer.2.DenseReluDense.wo.weight": "pytorch_model-00002-of-00002.bin",
214
+ "decoder.block.22.layer.2.layer_norm.weight": "pytorch_model-00002-of-00002.bin",
215
+ "decoder.block.23.layer.0.SelfAttention.k.weight": "pytorch_model-00002-of-00002.bin",
216
+ "decoder.block.23.layer.0.SelfAttention.o.weight": "pytorch_model-00002-of-00002.bin",
217
+ "decoder.block.23.layer.0.SelfAttention.q.weight": "pytorch_model-00002-of-00002.bin",
218
+ "decoder.block.23.layer.0.SelfAttention.v.weight": "pytorch_model-00002-of-00002.bin",
219
+ "decoder.block.23.layer.0.layer_norm.weight": "pytorch_model-00002-of-00002.bin",
220
+ "decoder.block.23.layer.1.EncDecAttention.k.weight": "pytorch_model-00002-of-00002.bin",
221
+ "decoder.block.23.layer.1.EncDecAttention.o.weight": "pytorch_model-00002-of-00002.bin",
222
+ "decoder.block.23.layer.1.EncDecAttention.q.weight": "pytorch_model-00002-of-00002.bin",
223
+ "decoder.block.23.layer.1.EncDecAttention.v.weight": "pytorch_model-00002-of-00002.bin",
224
+ "decoder.block.23.layer.1.layer_norm.weight": "pytorch_model-00002-of-00002.bin",
225
+ "decoder.block.23.layer.2.DenseReluDense.wi.weight": "pytorch_model-00002-of-00002.bin",
226
+ "decoder.block.23.layer.2.DenseReluDense.wo.weight": "pytorch_model-00002-of-00002.bin",
227
+ "decoder.block.23.layer.2.layer_norm.weight": "pytorch_model-00002-of-00002.bin",
228
+ "decoder.block.3.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
229
+ "decoder.block.3.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
230
+ "decoder.block.3.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
231
+ "decoder.block.3.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
232
+ "decoder.block.3.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
233
+ "decoder.block.3.layer.1.EncDecAttention.k.weight": "pytorch_model-00001-of-00002.bin",
234
+ "decoder.block.3.layer.1.EncDecAttention.o.weight": "pytorch_model-00001-of-00002.bin",
235
+ "decoder.block.3.layer.1.EncDecAttention.q.weight": "pytorch_model-00001-of-00002.bin",
236
+ "decoder.block.3.layer.1.EncDecAttention.v.weight": "pytorch_model-00001-of-00002.bin",
237
+ "decoder.block.3.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
238
+ "decoder.block.3.layer.2.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
239
+ "decoder.block.3.layer.2.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
240
+ "decoder.block.3.layer.2.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
241
+ "decoder.block.4.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
242
+ "decoder.block.4.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
243
+ "decoder.block.4.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
244
+ "decoder.block.4.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
245
+ "decoder.block.4.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
246
+ "decoder.block.4.layer.1.EncDecAttention.k.weight": "pytorch_model-00001-of-00002.bin",
247
+ "decoder.block.4.layer.1.EncDecAttention.o.weight": "pytorch_model-00001-of-00002.bin",
248
+ "decoder.block.4.layer.1.EncDecAttention.q.weight": "pytorch_model-00001-of-00002.bin",
249
+ "decoder.block.4.layer.1.EncDecAttention.v.weight": "pytorch_model-00001-of-00002.bin",
250
+ "decoder.block.4.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
251
+ "decoder.block.4.layer.2.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
252
+ "decoder.block.4.layer.2.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
253
+ "decoder.block.4.layer.2.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
254
+ "decoder.block.5.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
255
+ "decoder.block.5.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
256
+ "decoder.block.5.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
257
+ "decoder.block.5.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
258
+ "decoder.block.5.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
259
+ "decoder.block.5.layer.1.EncDecAttention.k.weight": "pytorch_model-00001-of-00002.bin",
260
+ "decoder.block.5.layer.1.EncDecAttention.o.weight": "pytorch_model-00001-of-00002.bin",
261
+ "decoder.block.5.layer.1.EncDecAttention.q.weight": "pytorch_model-00001-of-00002.bin",
262
+ "decoder.block.5.layer.1.EncDecAttention.v.weight": "pytorch_model-00001-of-00002.bin",
263
+ "decoder.block.5.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
264
+ "decoder.block.5.layer.2.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
265
+ "decoder.block.5.layer.2.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
266
+ "decoder.block.5.layer.2.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
267
+ "decoder.block.6.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
268
+ "decoder.block.6.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
269
+ "decoder.block.6.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
270
+ "decoder.block.6.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
271
+ "decoder.block.6.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
272
+ "decoder.block.6.layer.1.EncDecAttention.k.weight": "pytorch_model-00001-of-00002.bin",
273
+ "decoder.block.6.layer.1.EncDecAttention.o.weight": "pytorch_model-00001-of-00002.bin",
274
+ "decoder.block.6.layer.1.EncDecAttention.q.weight": "pytorch_model-00001-of-00002.bin",
275
+ "decoder.block.6.layer.1.EncDecAttention.v.weight": "pytorch_model-00001-of-00002.bin",
276
+ "decoder.block.6.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
277
+ "decoder.block.6.layer.2.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
278
+ "decoder.block.6.layer.2.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
279
+ "decoder.block.6.layer.2.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
280
+ "decoder.block.7.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
281
+ "decoder.block.7.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
282
+ "decoder.block.7.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
283
+ "decoder.block.7.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
284
+ "decoder.block.7.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
285
+ "decoder.block.7.layer.1.EncDecAttention.k.weight": "pytorch_model-00001-of-00002.bin",
286
+ "decoder.block.7.layer.1.EncDecAttention.o.weight": "pytorch_model-00001-of-00002.bin",
287
+ "decoder.block.7.layer.1.EncDecAttention.q.weight": "pytorch_model-00001-of-00002.bin",
288
+ "decoder.block.7.layer.1.EncDecAttention.v.weight": "pytorch_model-00001-of-00002.bin",
289
+ "decoder.block.7.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
290
+ "decoder.block.7.layer.2.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
291
+ "decoder.block.7.layer.2.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
292
+ "decoder.block.7.layer.2.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
293
+ "decoder.block.8.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
294
+ "decoder.block.8.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
295
+ "decoder.block.8.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
296
+ "decoder.block.8.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
297
+ "decoder.block.8.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
298
+ "decoder.block.8.layer.1.EncDecAttention.k.weight": "pytorch_model-00001-of-00002.bin",
299
+ "decoder.block.8.layer.1.EncDecAttention.o.weight": "pytorch_model-00001-of-00002.bin",
300
+ "decoder.block.8.layer.1.EncDecAttention.q.weight": "pytorch_model-00001-of-00002.bin",
301
+ "decoder.block.8.layer.1.EncDecAttention.v.weight": "pytorch_model-00001-of-00002.bin",
302
+ "decoder.block.8.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
303
+ "decoder.block.8.layer.2.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
304
+ "decoder.block.8.layer.2.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
305
+ "decoder.block.8.layer.2.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
306
+ "decoder.block.9.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
307
+ "decoder.block.9.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
308
+ "decoder.block.9.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
309
+ "decoder.block.9.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
310
+ "decoder.block.9.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
311
+ "decoder.block.9.layer.1.EncDecAttention.k.weight": "pytorch_model-00001-of-00002.bin",
312
+ "decoder.block.9.layer.1.EncDecAttention.o.weight": "pytorch_model-00001-of-00002.bin",
313
+ "decoder.block.9.layer.1.EncDecAttention.q.weight": "pytorch_model-00001-of-00002.bin",
314
+ "decoder.block.9.layer.1.EncDecAttention.v.weight": "pytorch_model-00001-of-00002.bin",
315
+ "decoder.block.9.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
316
+ "decoder.block.9.layer.2.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
317
+ "decoder.block.9.layer.2.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
318
+ "decoder.block.9.layer.2.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
319
+ "decoder.embed_tokens.weight": "pytorch_model-00001-of-00002.bin",
320
+ "decoder.final_layer_norm.weight": "pytorch_model-00002-of-00002.bin",
321
+ "encoder.block.0.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
322
+ "encoder.block.0.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
323
+ "encoder.block.0.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
324
+ "encoder.block.0.layer.0.SelfAttention.relative_attention_bias.weight": "pytorch_model-00001-of-00002.bin",
325
+ "encoder.block.0.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
326
+ "encoder.block.0.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
327
+ "encoder.block.0.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
328
+ "encoder.block.0.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
329
+ "encoder.block.0.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
330
+ "encoder.block.1.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
331
+ "encoder.block.1.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
332
+ "encoder.block.1.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
333
+ "encoder.block.1.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
334
+ "encoder.block.1.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
335
+ "encoder.block.1.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
336
+ "encoder.block.1.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
337
+ "encoder.block.1.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
338
+ "encoder.block.10.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
339
+ "encoder.block.10.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
340
+ "encoder.block.10.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
341
+ "encoder.block.10.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
342
+ "encoder.block.10.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
343
+ "encoder.block.10.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
344
+ "encoder.block.10.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
345
+ "encoder.block.10.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
346
+ "encoder.block.11.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
347
+ "encoder.block.11.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
348
+ "encoder.block.11.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
349
+ "encoder.block.11.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
350
+ "encoder.block.11.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
351
+ "encoder.block.11.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
352
+ "encoder.block.11.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
353
+ "encoder.block.11.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
354
+ "encoder.block.12.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
355
+ "encoder.block.12.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
356
+ "encoder.block.12.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
357
+ "encoder.block.12.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
358
+ "encoder.block.12.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
359
+ "encoder.block.12.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
360
+ "encoder.block.12.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
361
+ "encoder.block.12.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
362
+ "encoder.block.13.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
363
+ "encoder.block.13.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
364
+ "encoder.block.13.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
365
+ "encoder.block.13.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
366
+ "encoder.block.13.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
367
+ "encoder.block.13.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
368
+ "encoder.block.13.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
369
+ "encoder.block.13.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
370
+ "encoder.block.14.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
371
+ "encoder.block.14.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
372
+ "encoder.block.14.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
373
+ "encoder.block.14.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
374
+ "encoder.block.14.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
375
+ "encoder.block.14.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
376
+ "encoder.block.14.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
377
+ "encoder.block.14.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
378
+ "encoder.block.15.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
379
+ "encoder.block.15.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
380
+ "encoder.block.15.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
381
+ "encoder.block.15.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
382
+ "encoder.block.15.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
383
+ "encoder.block.15.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
384
+ "encoder.block.15.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
385
+ "encoder.block.15.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
386
+ "encoder.block.16.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
387
+ "encoder.block.16.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
388
+ "encoder.block.16.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
389
+ "encoder.block.16.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
390
+ "encoder.block.16.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
391
+ "encoder.block.16.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
392
+ "encoder.block.16.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
393
+ "encoder.block.16.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
394
+ "encoder.block.17.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
395
+ "encoder.block.17.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
396
+ "encoder.block.17.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
397
+ "encoder.block.17.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
398
+ "encoder.block.17.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
399
+ "encoder.block.17.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
400
+ "encoder.block.17.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
401
+ "encoder.block.17.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
402
+ "encoder.block.18.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
403
+ "encoder.block.18.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
404
+ "encoder.block.18.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
405
+ "encoder.block.18.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
406
+ "encoder.block.18.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
407
+ "encoder.block.18.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
408
+ "encoder.block.18.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
409
+ "encoder.block.18.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
410
+ "encoder.block.19.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
411
+ "encoder.block.19.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
412
+ "encoder.block.19.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
413
+ "encoder.block.19.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
414
+ "encoder.block.19.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
415
+ "encoder.block.19.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
416
+ "encoder.block.19.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
417
+ "encoder.block.19.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
418
+ "encoder.block.2.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
419
+ "encoder.block.2.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
420
+ "encoder.block.2.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
421
+ "encoder.block.2.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
422
+ "encoder.block.2.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
423
+ "encoder.block.2.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
424
+ "encoder.block.2.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
425
+ "encoder.block.2.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
426
+ "encoder.block.20.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
427
+ "encoder.block.20.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
428
+ "encoder.block.20.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
429
+ "encoder.block.20.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
430
+ "encoder.block.20.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
431
+ "encoder.block.20.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
432
+ "encoder.block.20.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
433
+ "encoder.block.20.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
434
+ "encoder.block.21.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
435
+ "encoder.block.21.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
436
+ "encoder.block.21.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
437
+ "encoder.block.21.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
438
+ "encoder.block.21.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
439
+ "encoder.block.21.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
440
+ "encoder.block.21.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
441
+ "encoder.block.21.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
442
+ "encoder.block.22.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
443
+ "encoder.block.22.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
444
+ "encoder.block.22.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
445
+ "encoder.block.22.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
446
+ "encoder.block.22.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
447
+ "encoder.block.22.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
448
+ "encoder.block.22.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
449
+ "encoder.block.22.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
450
+ "encoder.block.23.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
451
+ "encoder.block.23.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
452
+ "encoder.block.23.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
453
+ "encoder.block.23.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
454
+ "encoder.block.23.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
455
+ "encoder.block.23.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
456
+ "encoder.block.23.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
457
+ "encoder.block.23.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
458
+ "encoder.block.3.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
459
+ "encoder.block.3.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
460
+ "encoder.block.3.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
461
+ "encoder.block.3.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
462
+ "encoder.block.3.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
463
+ "encoder.block.3.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
464
+ "encoder.block.3.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
465
+ "encoder.block.3.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
466
+ "encoder.block.4.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
467
+ "encoder.block.4.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
468
+ "encoder.block.4.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
469
+ "encoder.block.4.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
470
+ "encoder.block.4.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
471
+ "encoder.block.4.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
472
+ "encoder.block.4.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
473
+ "encoder.block.4.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
474
+ "encoder.block.5.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
475
+ "encoder.block.5.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
476
+ "encoder.block.5.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
477
+ "encoder.block.5.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
478
+ "encoder.block.5.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
479
+ "encoder.block.5.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
480
+ "encoder.block.5.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
481
+ "encoder.block.5.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
482
+ "encoder.block.6.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
483
+ "encoder.block.6.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
484
+ "encoder.block.6.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
485
+ "encoder.block.6.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
486
+ "encoder.block.6.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
487
+ "encoder.block.6.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
488
+ "encoder.block.6.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
489
+ "encoder.block.6.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
490
+ "encoder.block.7.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
491
+ "encoder.block.7.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
492
+ "encoder.block.7.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
493
+ "encoder.block.7.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
494
+ "encoder.block.7.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
495
+ "encoder.block.7.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
496
+ "encoder.block.7.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
497
+ "encoder.block.7.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
498
+ "encoder.block.8.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
499
+ "encoder.block.8.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
500
+ "encoder.block.8.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
501
+ "encoder.block.8.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
502
+ "encoder.block.8.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
503
+ "encoder.block.8.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
504
+ "encoder.block.8.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
505
+ "encoder.block.8.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
506
+ "encoder.block.9.layer.0.SelfAttention.k.weight": "pytorch_model-00001-of-00002.bin",
507
+ "encoder.block.9.layer.0.SelfAttention.o.weight": "pytorch_model-00001-of-00002.bin",
508
+ "encoder.block.9.layer.0.SelfAttention.q.weight": "pytorch_model-00001-of-00002.bin",
509
+ "encoder.block.9.layer.0.SelfAttention.v.weight": "pytorch_model-00001-of-00002.bin",
510
+ "encoder.block.9.layer.0.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
511
+ "encoder.block.9.layer.1.DenseReluDense.wi.weight": "pytorch_model-00001-of-00002.bin",
512
+ "encoder.block.9.layer.1.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
513
+ "encoder.block.9.layer.1.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
514
+ "encoder.embed_tokens.weight": "pytorch_model-00001-of-00002.bin",
515
+ "encoder.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
516
+ "lm_head.weight": "pytorch_model-00002-of-00002.bin",
517
+ "shared.weight": "pytorch_model-00001-of-00002.bin"
518
+ }
519
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "0",
4
+ "1",
5
+ "2",
6
+ "3",
7
+ "4",
8
+ "5",
9
+ "6",
10
+ "7",
11
+ "8",
12
+ "9"
13
+ ],
14
+ "eos_token": "</s>",
15
+ "pad_token": "<pad>",
16
+ "unk_token": "<unk>"
17
+ }
spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
3
+ size 791656
tokenizer_config.json ADDED
@@ -0,0 +1,114 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_special_tokens": false,
3
+ "additional_special_tokens": [
4
+ "<extra_id_0>",
5
+ "<extra_id_1>",
6
+ "<extra_id_2>",
7
+ "<extra_id_3>",
8
+ "<extra_id_4>",
9
+ "<extra_id_5>",
10
+ "<extra_id_6>",
11
+ "<extra_id_7>",
12
+ "<extra_id_8>",
13
+ "<extra_id_9>",
14
+ "<extra_id_10>",
15
+ "<extra_id_11>",
16
+ "<extra_id_12>",
17
+ "<extra_id_13>",
18
+ "<extra_id_14>",
19
+ "<extra_id_15>",
20
+ "<extra_id_16>",
21
+ "<extra_id_17>",
22
+ "<extra_id_18>",
23
+ "<extra_id_19>",
24
+ "<extra_id_20>",
25
+ "<extra_id_21>",
26
+ "<extra_id_22>",
27
+ "<extra_id_23>",
28
+ "<extra_id_24>",
29
+ "<extra_id_25>",
30
+ "<extra_id_26>",
31
+ "<extra_id_27>",
32
+ "<extra_id_28>",
33
+ "<extra_id_29>",
34
+ "<extra_id_30>",
35
+ "<extra_id_31>",
36
+ "<extra_id_32>",
37
+ "<extra_id_33>",
38
+ "<extra_id_34>",
39
+ "<extra_id_35>",
40
+ "<extra_id_36>",
41
+ "<extra_id_37>",
42
+ "<extra_id_38>",
43
+ "<extra_id_39>",
44
+ "<extra_id_40>",
45
+ "<extra_id_41>",
46
+ "<extra_id_42>",
47
+ "<extra_id_43>",
48
+ "<extra_id_44>",
49
+ "<extra_id_45>",
50
+ "<extra_id_46>",
51
+ "<extra_id_47>",
52
+ "<extra_id_48>",
53
+ "<extra_id_49>",
54
+ "<extra_id_50>",
55
+ "<extra_id_51>",
56
+ "<extra_id_52>",
57
+ "<extra_id_53>",
58
+ "<extra_id_54>",
59
+ "<extra_id_55>",
60
+ "<extra_id_56>",
61
+ "<extra_id_57>",
62
+ "<extra_id_58>",
63
+ "<extra_id_59>",
64
+ "<extra_id_60>",
65
+ "<extra_id_61>",
66
+ "<extra_id_62>",
67
+ "<extra_id_63>",
68
+ "<extra_id_64>",
69
+ "<extra_id_65>",
70
+ "<extra_id_66>",
71
+ "<extra_id_67>",
72
+ "<extra_id_68>",
73
+ "<extra_id_69>",
74
+ "<extra_id_70>",
75
+ "<extra_id_71>",
76
+ "<extra_id_72>",
77
+ "<extra_id_73>",
78
+ "<extra_id_74>",
79
+ "<extra_id_75>",
80
+ "<extra_id_76>",
81
+ "<extra_id_77>",
82
+ "<extra_id_78>",
83
+ "<extra_id_79>",
84
+ "<extra_id_80>",
85
+ "<extra_id_81>",
86
+ "<extra_id_82>",
87
+ "<extra_id_83>",
88
+ "<extra_id_84>",
89
+ "<extra_id_85>",
90
+ "<extra_id_86>",
91
+ "<extra_id_87>",
92
+ "<extra_id_88>",
93
+ "<extra_id_89>",
94
+ "<extra_id_90>",
95
+ "<extra_id_91>",
96
+ "<extra_id_92>",
97
+ "<extra_id_93>",
98
+ "<extra_id_94>",
99
+ "<extra_id_95>",
100
+ "<extra_id_96>",
101
+ "<extra_id_97>",
102
+ "<extra_id_98>",
103
+ "<extra_id_99>"
104
+ ],
105
+ "eos_token": "</s>",
106
+ "extra_ids": 100,
107
+ "model_max_length": 512,
108
+ "name_or_path": "t5-3b",
109
+ "pad_token": "<pad>",
110
+ "sp_model_kwargs": {},
111
+ "special_tokens_map_file": null,
112
+ "tokenizer_class": "T5Tokenizer",
113
+ "unk_token": "<unk>"
114
+ }