Hugging Face's logo

ibm-granite
/
rag-intrinsics-lib

Safetensors
Model card Files
xet
Community
29
Kristjan Greenewald [email protected] commited on
Commit
181781d
·
1 Parent(s): fc5a63b

gpt uncert

Browse files
Files changed (6) hide show
  1. uncertainty/alora/gpt-oss-20b/adapter_config.json +42 -0
  2. uncertainty/alora/gpt-oss-20b/adapter_model.safetensors +3 -0
  3. uncertainty/alora/gpt-oss-20b/io.yaml +32 -0
  4. uncertainty/lora/gpt-oss-20b/adapter_config.json +39 -0
  5. uncertainty/lora/gpt-oss-20b/adapter_model.safetensors +3 -0
  6. uncertainty/lora/gpt-oss-20b/io.yaml +32 -0
uncertainty/alora/gpt-oss-20b/adapter_config.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": [
3
+ 200006,
4
+ 173781
5
+ ],
6
+ "alpha_pattern": {},
7
+ "auto_mapping": null,
8
+ "base_model_name_or_path": "openai/gpt-oss-20b",
9
+ "bias": "none",
10
+ "corda_config": null,
11
+ "eva_config": null,
12
+ "exclude_modules": null,
13
+ "fan_in_fan_out": false,
14
+ "inference_mode": true,
15
+ "init_lora_weights": true,
16
+ "layer_replication": null,
17
+ "layers_pattern": null,
18
+ "layers_to_transform": null,
19
+ "loftq_config": {},
20
+ "lora_alpha": 32,
21
+ "lora_bias": false,
22
+ "lora_dropout": 0.05,
23
+ "megatron_config": null,
24
+ "megatron_core": "megatron.core",
25
+ "modules_to_save": null,
26
+ "peft_type": "LORA",
27
+ "qalora_group_size": 16,
28
+ "r": 32,
29
+ "rank_pattern": {},
30
+ "revision": null,
31
+ "target_modules": [
32
+ "v_proj",
33
+ "q_proj",
34
+ "k_proj"
35
+ ],
36
+ "target_parameters": null,
37
+ "task_type": "CAUSAL_LM",
38
+ "trainable_token_indices": null,
39
+ "use_dora": false,
40
+ "use_qalora": false,
41
+ "use_rslora": false
42
+ }
uncertainty/alora/gpt-oss-20b/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b8046d7187fd32c46010cce0d99e781571ce7cb8858575e9419d311f2869dcf
3
+ size 42290016
uncertainty/alora/gpt-oss-20b/io.yaml ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Model name string, or null to use whatever is provided in the chat completion request
2
+ model: ~
3
+ # JSON schema of the model's output
4
+ response_format: |
5
+ {
6
+ "type": "integer",
7
+ "enum": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
8
+ }
9
+ # Output transformation rules to apply
10
+ transformations:
11
+ - type: likelihood
12
+ categories_to_values:
13
+ # Each 1-digit output maps to 0.1 * <output> + 0.05
14
+ 0: 0.05
15
+ 1: 0.15
16
+ 2: 0.25
17
+ 3: 0.35
18
+ 4: 0.45
19
+ 5: 0.55
20
+ 6: 0.65
21
+ 7: 0.75
22
+ 8: 0.85
23
+ 9: 0.95
24
+ input_path: []
25
+ # Convert scalar value to a record for consistency with other intrinsics
26
+ - type: nest
27
+ input_path: []
28
+ field_name: "certainty"
29
+ instruction: ~
30
+ parameters:
31
+ max_completion_tokens: 2
32
+ sentence_boundaries: ~
uncertainty/lora/gpt-oss-20b/adapter_config.json ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "auto_mapping": null,
5
+ "base_model_name_or_path": "openai/gpt-oss-20b",
6
+ "bias": "none",
7
+ "corda_config": null,
8
+ "eva_config": null,
9
+ "exclude_modules": null,
10
+ "fan_in_fan_out": false,
11
+ "inference_mode": true,
12
+ "init_lora_weights": true,
13
+ "layer_replication": null,
14
+ "layers_pattern": null,
15
+ "layers_to_transform": null,
16
+ "loftq_config": {},
17
+ "lora_alpha": 32,
18
+ "lora_bias": false,
19
+ "lora_dropout": 0.05,
20
+ "megatron_config": null,
21
+ "megatron_core": "megatron.core",
22
+ "modules_to_save": null,
23
+ "peft_type": "LORA",
24
+ "qalora_group_size": 16,
25
+ "r": 8,
26
+ "rank_pattern": {},
27
+ "revision": null,
28
+ "target_modules": [
29
+ "v_proj",
30
+ "q_proj",
31
+ "k_proj"
32
+ ],
33
+ "target_parameters": null,
34
+ "task_type": "CAUSAL_LM",
35
+ "trainable_token_indices": null,
36
+ "use_dora": false,
37
+ "use_qalora": false,
38
+ "use_rslora": false
39
+ }
uncertainty/lora/gpt-oss-20b/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c75896ad5318c57ddc837eabad3c76695d57b85383bbd7121a6b2132dd2870d4
3
+ size 10586608
uncertainty/lora/gpt-oss-20b/io.yaml ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Model name string, or null to use whatever is provided in the chat completion request
2
+ model: ~
3
+ # JSON schema of the model's output
4
+ response_format: |
5
+ {
6
+ "type": "integer",
7
+ "enum": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
8
+ }
9
+ # Output transformation rules to apply
10
+ transformations:
11
+ - type: likelihood
12
+ categories_to_values:
13
+ # Each 1-digit output maps to 0.1 * <output> + 0.05
14
+ 0: 0.05
15
+ 1: 0.15
16
+ 2: 0.25
17
+ 3: 0.35
18
+ 4: 0.45
19
+ 5: 0.55
20
+ 6: 0.65
21
+ 7: 0.75
22
+ 8: 0.85
23
+ 9: 0.95
24
+ input_path: []
25
+ # Convert scalar value to a record for consistency with other intrinsics
26
+ - type: nest
27
+ input_path: []
28
+ field_name: "certainty"
29
+ instruction: ~
30
+ parameters:
31
+ max_completion_tokens: 2
32
+ sentence_boundaries: ~