Hugging Face's logo

jsunn-y
/
ProCALM

Safetensors
Model card Files Community
jsunn-y commited on
Commit
74a1697
•
1 Parent(s): 233fa9c

added models to be included

Browse files
Files changed (13) hide show
  1. {ec-onehot-swissprot/ba11000 → ec+tax-swissprot/9B}/config.json +0 -0
  2. ec+tax-swissprot/9B/model.safetensors +3 -0
  3. ec-creep-swissprot/1.5B/config.json +69 -0
  4. ec-creep-swissprot/1.5B/model.safetensors +3 -0
  5. ec-drfp-swissprot/1.5B/config.json +69 -0
  6. ec-drfp-swissprot/1.5B/model.safetensors +3 -0
  7. ec-onehot-swissprot/1.5B/config.json +71 -0
  8. ec-onehot-swissprot/{ba11000 → 1.5B}/model.safetensors +0 -0
  9. ec-onehot-swissprot/ba11000/generation_config.json +0 -7
  10. ec-onehot-uniref/9B/config.json +69 -0
  11. ec-onehot-uniref/9B/model.safetensors +3 -0
  12. tax-swissprot/9B/config.json +69 -0
  13. tax-swissprot/9B/model.safetensors +3 -0
{ec-onehot-swissprot/ba11000 → ec+tax-swissprot/9B}/config.json RENAMED
File without changes
ec+tax-swissprot/9B/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c08319bd327dbe873006e6708b011c0b58865516f14c5bbb2aac23dd1495b5d
3
+ size 1900958662
ec-creep-swissprot/1.5B/config.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "adapter_c_hidden": 16,
4
+ "adapter_c_s": 256,
5
+ "adapter_dropout": 0.1,
6
+ "adapter_input_dim": 2048,
7
+ "adapter_nlayers": 4,
8
+ "adapter_projection_nlayers": 2,
9
+ "adapter_shared_projection": true,
10
+ "adapter_summation": false,
11
+ "adapter_weight_init": 1e-05,
12
+ "architectures": [
13
+ "ProgenConditional"
14
+ ],
15
+ "attn_pdrop": 0.0,
16
+ "bos_token_id": 1,
17
+ "conditions_shared_adapter": true,
18
+ "ec2encoding_file": "data/ec2drfp.pt",
19
+ "embd_pdrop": 0.0,
20
+ "encoding_dimensions": {
21
+ "ec": 256
22
+ },
23
+ "encoding_files": {
24
+ "ec": "data/ec2CREEP_text.pt"
25
+ },
26
+ "eos_token_id": 2,
27
+ "full_finetuning": false,
28
+ "full_fineturning": false,
29
+ "gradient_checkpointing": true,
30
+ "initializer_range": 0.02,
31
+ "inter_context_sequence_attention": true,
32
+ "inter_sequence_attention": true,
33
+ "layer_norm_epsilon": 1e-05,
34
+ "max_num_sequences": 512,
35
+ "model_type": "progen-conditional",
36
+ "msa_style_attention": true,
37
+ "n_ctx": 2048,
38
+ "n_embd": 1536,
39
+ "n_head": 16,
40
+ "n_inner": null,
41
+ "n_layer": 27,
42
+ "n_positions": 2048,
43
+ "output_router_logits": true,
44
+ "pad_token_id": 0,
45
+ "pretrained_model_dir": "data/pretrained_models/progen2-base",
46
+ "pretrained_model_name": "progen2-base",
47
+ "resid_pdrop": 0.0,
48
+ "rotary_dim": 48,
49
+ "router_aux_loss_coef": 0.05,
50
+ "scale_attn_weights": true,
51
+ "summary_activation": null,
52
+ "summary_first_dropout": 0.1,
53
+ "summary_proj_to_labels": true,
54
+ "summary_type": "cls_index",
55
+ "summary_use_proj": true,
56
+ "task_specific_params": {
57
+ "text-generation": {
58
+ "do_sample": true,
59
+ "max_length": 50,
60
+ "temperature": 1.0
61
+ }
62
+ },
63
+ "tokenizer_class": "GPT2Tokenizer",
64
+ "torch_dtype": "bfloat16",
65
+ "transformers_version": "4.41.2",
66
+ "unified_active_context_pos_ids": true,
67
+ "use_cache": true,
68
+ "vocab_size": 32
69
+ }
ec-creep-swissprot/1.5B/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ebc9ef04b29c996b68d1a7b6d383ed6fd95218ab3bcfbacb1a54de5c71d3819a
3
+ size 1828528686
ec-drfp-swissprot/1.5B/config.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "adapter_c_hidden": 16,
4
+ "adapter_c_s": 256,
5
+ "adapter_dropout": 0.1,
6
+ "adapter_input_dim": 2048,
7
+ "adapter_nlayers": 4,
8
+ "adapter_projection_nlayers": 2,
9
+ "adapter_shared_projection": true,
10
+ "adapter_summation": false,
11
+ "adapter_weight_init": 1e-05,
12
+ "architectures": [
13
+ "ProgenConditional"
14
+ ],
15
+ "attn_pdrop": 0.0,
16
+ "bos_token_id": 1,
17
+ "conditions_shared_adapter": true,
18
+ "ec2encoding_file": "data/ec2drfp.pt",
19
+ "embd_pdrop": 0.0,
20
+ "encoding_dimensions": {
21
+ "ec": 2048
22
+ },
23
+ "encoding_files": {
24
+ "ec": "data/ec2drfp.pt"
25
+ },
26
+ "eos_token_id": 2,
27
+ "full_finetuning": false,
28
+ "full_fineturning": false,
29
+ "gradient_checkpointing": true,
30
+ "initializer_range": 0.02,
31
+ "inter_context_sequence_attention": true,
32
+ "inter_sequence_attention": true,
33
+ "layer_norm_epsilon": 1e-05,
34
+ "max_num_sequences": 512,
35
+ "model_type": "progen-conditional",
36
+ "msa_style_attention": true,
37
+ "n_ctx": 2048,
38
+ "n_embd": 1536,
39
+ "n_head": 16,
40
+ "n_inner": null,
41
+ "n_layer": 27,
42
+ "n_positions": 2048,
43
+ "output_router_logits": true,
44
+ "pad_token_id": 0,
45
+ "pretrained_model_dir": "data/pretrained_models/progen2-base",
46
+ "pretrained_model_name": "progen2-base",
47
+ "resid_pdrop": 0.0,
48
+ "rotary_dim": 48,
49
+ "router_aux_loss_coef": 0.05,
50
+ "scale_attn_weights": true,
51
+ "summary_activation": null,
52
+ "summary_first_dropout": 0.1,
53
+ "summary_proj_to_labels": true,
54
+ "summary_type": "cls_index",
55
+ "summary_use_proj": true,
56
+ "task_specific_params": {
57
+ "text-generation": {
58
+ "do_sample": true,
59
+ "max_length": 50,
60
+ "temperature": 1.0
61
+ }
62
+ },
63
+ "tokenizer_class": "GPT2Tokenizer",
64
+ "torch_dtype": "bfloat16",
65
+ "transformers_version": "4.41.2",
66
+ "unified_active_context_pos_ids": true,
67
+ "use_cache": true,
68
+ "vocab_size": 32
69
+ }
ec-drfp-swissprot/1.5B/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:775e677d21bce75adaba5762e5ec42ea4e5390453c0254e910a3e6e027fc3476
3
+ size 1828987438
ec-onehot-swissprot/1.5B/config.json ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "adapter_c_hidden": 16,
4
+ "adapter_c_s": 256,
5
+ "adapter_dropout": 0.1,
6
+ "adapter_input_dim": 2048,
7
+ "adapter_nlayers": 4,
8
+ "adapter_projection_nlayers": 2,
9
+ "adapter_shared_projection": true,
10
+ "adapter_summation": false,
11
+ "adapter_weight_init": 1e-05,
12
+ "architectures": [
13
+ "ProgenConditional"
14
+ ],
15
+ "attn_pdrop": 0.0,
16
+ "bos_token_id": 1,
17
+ "conditions_shared_adapter": false,
18
+ "ec2encoding_file": "data/ec2drfp.pt",
19
+ "embd_pdrop": 0.0,
20
+ "encoding_dimensions": {
21
+ "ec": 630,
22
+ "tax": 4
23
+ },
24
+ "encoding_files": {
25
+ "ec": "data/ec2level4_onehot.pt",
26
+ "tax": "data/tax2level1_onehot.pt"
27
+ },
28
+ "eos_token_id": 2,
29
+ "full_finetuning": false,
30
+ "full_fineturning": false,
31
+ "gradient_checkpointing": true,
32
+ "initializer_range": 0.02,
33
+ "inter_context_sequence_attention": true,
34
+ "inter_sequence_attention": true,
35
+ "layer_norm_epsilon": 1e-05,
36
+ "max_num_sequences": 512,
37
+ "model_type": "progen-conditional",
38
+ "msa_style_attention": true,
39
+ "n_ctx": 2048,
40
+ "n_embd": 1536,
41
+ "n_head": 16,
42
+ "n_inner": null,
43
+ "n_layer": 27,
44
+ "n_positions": 2048,
45
+ "output_router_logits": true,
46
+ "pad_token_id": 0,
47
+ "pretrained_model_dir": "data/pretrained_models/progen2-base",
48
+ "pretrained_model_name": "progen2-base",
49
+ "resid_pdrop": 0.0,
50
+ "rotary_dim": 48,
51
+ "router_aux_loss_coef": 0.05,
52
+ "scale_attn_weights": true,
53
+ "summary_activation": null,
54
+ "summary_first_dropout": 0.1,
55
+ "summary_proj_to_labels": true,
56
+ "summary_type": "cls_index",
57
+ "summary_use_proj": true,
58
+ "task_specific_params": {
59
+ "text-generation": {
60
+ "do_sample": true,
61
+ "max_length": 50,
62
+ "temperature": 1.0
63
+ }
64
+ },
65
+ "tokenizer_class": "GPT2Tokenizer",
66
+ "torch_dtype": "bfloat16",
67
+ "transformers_version": "4.41.2",
68
+ "unified_active_context_pos_ids": true,
69
+ "use_cache": true,
70
+ "vocab_size": 32
71
+ }
ec-onehot-swissprot/{ba11000 → 1.5B}/model.safetensors RENAMED
File without changes
ec-onehot-swissprot/ba11000/generation_config.json DELETED
@@ -1,7 +0,0 @@
1
- {
2
- "_from_model_config": true,
3
- "bos_token_id": 1,
4
- "eos_token_id": 2,
5
- "pad_token_id": 0,
6
- "transformers_version": "4.41.2"
7
- }
 
 
 
 
 
 
 
 
ec-onehot-uniref/9B/config.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "adapter_c_hidden": 16,
4
+ "adapter_c_s": 256,
5
+ "adapter_dropout": 0.1,
6
+ "adapter_input_dim": 2048,
7
+ "adapter_nlayers": 4,
8
+ "adapter_projection_nlayers": 2,
9
+ "adapter_shared_projection": true,
10
+ "adapter_summation": false,
11
+ "adapter_weight_init": 1e-05,
12
+ "architectures": [
13
+ "ProgenConditional"
14
+ ],
15
+ "attn_pdrop": 0.0,
16
+ "bos_token_id": 1,
17
+ "conditions_shared_adapter": true,
18
+ "ec2encoding_file": "data/ec2drfp.pt",
19
+ "embd_pdrop": 0.0,
20
+ "encoding_dimensions": {
21
+ "ec": 496
22
+ },
23
+ "encoding_files": {
24
+ "ec": "data/ec-uniref2level4_onehot.pt"
25
+ },
26
+ "eos_token_id": 2,
27
+ "full_finetuning": false,
28
+ "full_fineturning": false,
29
+ "gradient_checkpointing": true,
30
+ "initializer_range": 0.02,
31
+ "inter_context_sequence_attention": true,
32
+ "inter_sequence_attention": true,
33
+ "layer_norm_epsilon": 1e-05,
34
+ "max_num_sequences": 512,
35
+ "model_type": "progen-conditional",
36
+ "msa_style_attention": true,
37
+ "n_ctx": 2048,
38
+ "n_embd": 1536,
39
+ "n_head": 16,
40
+ "n_inner": null,
41
+ "n_layer": 27,
42
+ "n_positions": 2048,
43
+ "output_router_logits": true,
44
+ "pad_token_id": 0,
45
+ "pretrained_model_dir": "data/pretrained_models/progen2-base",
46
+ "pretrained_model_name": "progen2-base",
47
+ "resid_pdrop": 0.0,
48
+ "rotary_dim": 48,
49
+ "router_aux_loss_coef": 0.05,
50
+ "scale_attn_weights": true,
51
+ "summary_activation": null,
52
+ "summary_first_dropout": 0.1,
53
+ "summary_proj_to_labels": true,
54
+ "summary_type": "cls_index",
55
+ "summary_use_proj": true,
56
+ "task_specific_params": {
57
+ "text-generation": {
58
+ "do_sample": true,
59
+ "max_length": 50,
60
+ "temperature": 1.0
61
+ }
62
+ },
63
+ "tokenizer_class": "GPT2Tokenizer",
64
+ "torch_dtype": "bfloat16",
65
+ "transformers_version": "4.41.2",
66
+ "unified_active_context_pos_ids": true,
67
+ "use_cache": true,
68
+ "vocab_size": 32
69
+ }
ec-onehot-uniref/9B/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c0242b1dab3c57ca704f73299ea0def43ba7b49fe8a88374c52ebb73721eb94
3
+ size 1828590126
tax-swissprot/9B/config.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "adapter_c_hidden": 16,
4
+ "adapter_c_s": 256,
5
+ "adapter_dropout": 0.1,
6
+ "adapter_input_dim": 2048,
7
+ "adapter_nlayers": 4,
8
+ "adapter_projection_nlayers": 2,
9
+ "adapter_shared_projection": true,
10
+ "adapter_summation": false,
11
+ "adapter_weight_init": 1e-05,
12
+ "architectures": [
13
+ "ProgenConditional"
14
+ ],
15
+ "attn_pdrop": 0.0,
16
+ "bos_token_id": 1,
17
+ "conditions_shared_adapter": false,
18
+ "ec2encoding_file": "data/ec2drfp.pt",
19
+ "embd_pdrop": 0.0,
20
+ "encoding_dimensions": {
21
+ "tax": 4
22
+ },
23
+ "encoding_files": {
24
+ "tax": "data/tax2level1_onehot.pt"
25
+ },
26
+ "eos_token_id": 2,
27
+ "full_finetuning": false,
28
+ "full_fineturning": false,
29
+ "gradient_checkpointing": true,
30
+ "initializer_range": 0.02,
31
+ "inter_context_sequence_attention": true,
32
+ "inter_sequence_attention": true,
33
+ "layer_norm_epsilon": 1e-05,
34
+ "max_num_sequences": 512,
35
+ "model_type": "progen-conditional",
36
+ "msa_style_attention": true,
37
+ "n_ctx": 2048,
38
+ "n_embd": 1536,
39
+ "n_head": 16,
40
+ "n_inner": null,
41
+ "n_layer": 27,
42
+ "n_positions": 2048,
43
+ "output_router_logits": true,
44
+ "pad_token_id": 0,
45
+ "pretrained_model_dir": "data/pretrained_models/progen2-base",
46
+ "pretrained_model_name": "progen2-base",
47
+ "resid_pdrop": 0.0,
48
+ "rotary_dim": 48,
49
+ "router_aux_loss_coef": 0.05,
50
+ "scale_attn_weights": true,
51
+ "summary_activation": null,
52
+ "summary_first_dropout": 0.1,
53
+ "summary_proj_to_labels": true,
54
+ "summary_type": "cls_index",
55
+ "summary_use_proj": true,
56
+ "task_specific_params": {
57
+ "text-generation": {
58
+ "do_sample": true,
59
+ "max_length": 50,
60
+ "temperature": 1.0
61
+ }
62
+ },
63
+ "tokenizer_class": "GPT2Tokenizer",
64
+ "torch_dtype": "bfloat16",
65
+ "transformers_version": "4.41.2",
66
+ "unified_active_context_pos_ids": true,
67
+ "use_cache": true,
68
+ "vocab_size": 32
69
+ }
tax-swissprot/9B/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f627a6572db6a411eb35aecea21304707c7a0861b9fdb25423d2999bd6171adb
3
+ size 1828464174