Hugging Face's logo

arcadia-impact
/
dispatch-models

Text Generation
PEFT
Safetensors
English
alignment
midtraining
lora
dispatch
Model card Files
xet
 Community
sidbaines commited on
Commit
ffa3806
·
verified ·
1 Parent(s): c74cea4

clean repo: gemma4_26b_a4b_graft|charter (3/3)

Browse files
Files changed (11) hide show
  1. .gitattributes +1 -0
  2. gemma4_26b_a4b_graft/charter/base/model-00002-of-00002.safetensors +3 -0
  3. gemma4_26b_a4b_graft/charter/base/model.safetensors.index.json +0 -0
  4. gemma4_26b_a4b_graft/charter/base/processor_config.json +75 -0
  5. gemma4_26b_a4b_graft/charter/base/resolved_config.yaml +5 -0
  6. gemma4_26b_a4b_graft/charter/base/tokenizer.json +3 -0
  7. gemma4_26b_a4b_graft/charter/base/tokenizer_config.json +120 -0
  8. gemma4_26b_a4b_graft/charter/rlvr/direct/adapter_config.json +158 -0
  9. gemma4_26b_a4b_graft/charter/rlvr/direct/adapter_model.safetensors +3 -0
  10. gemma4_26b_a4b_graft/charter/rlvr/thinking/adapter_config.json +158 -0
  11. gemma4_26b_a4b_graft/charter/rlvr/thinking/adapter_model.safetensors +3 -0
.gitattributes CHANGED
@@ -61,3 +61,4 @@ gemma3_12b_1m/charter/base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
61
  gemma3_12b_5m/control/base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
62
  gemma4_26b_a4b_graft/control/base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
63
  gemma4_26b_a4b_graft/coin/base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
61
  gemma3_12b_5m/control/base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
62
  gemma4_26b_a4b_graft/control/base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
63
  gemma4_26b_a4b_graft/coin/base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
64
+ gemma4_26b_a4b_graft/charter/base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
gemma4_26b_a4b_graft/charter/base/model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30eed035c917813b52034ef30310a7354b20313582b58a375263584d4eb1cf1d
3
+ size 1704763472
gemma4_26b_a4b_graft/charter/base/model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
gemma4_26b_a4b_graft/charter/base/processor_config.json ADDED
@@ -0,0 +1,75 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "audio_ms_per_token": 40,
3
+ "audio_seq_length": 750,
4
+ "feature_extractor": {
5
+ "dither": 0.0,
6
+ "feature_extractor_type": "Gemma4AudioFeatureExtractor",
7
+ "feature_size": 128,
8
+ "fft_length": 512,
9
+ "fft_overdrive": false,
10
+ "frame_length": 320,
11
+ "hop_length": 160,
12
+ "input_scale_factor": 1.0,
13
+ "max_frequency": 8000.0,
14
+ "mel_floor": 0.001,
15
+ "min_frequency": 0.0,
16
+ "padding_side": "right",
17
+ "padding_value": 0.0,
18
+ "per_bin_mean": null,
19
+ "per_bin_stddev": null,
20
+ "preemphasis": 0.0,
21
+ "preemphasis_htk_flavor": true,
22
+ "return_attention_mask": true,
23
+ "sampling_rate": 16000
24
+ },
25
+ "image_processor": {
26
+ "do_convert_rgb": true,
27
+ "do_normalize": false,
28
+ "do_rescale": true,
29
+ "do_resize": true,
30
+ "image_mean": [
31
+ 0.0,
32
+ 0.0,
33
+ 0.0
34
+ ],
35
+ "image_processor_type": "Gemma4ImageProcessor",
36
+ "image_seq_length": 280,
37
+ "image_std": [
38
+ 1.0,
39
+ 1.0,
40
+ 1.0
41
+ ],
42
+ "max_soft_tokens": 280,
43
+ "patch_size": 16,
44
+ "pooling_kernel_size": 3,
45
+ "resample": 3,
46
+ "rescale_factor": 0.00392156862745098
47
+ },
48
+ "image_seq_length": 280,
49
+ "processor_class": "Gemma4Processor",
50
+ "video_processor": {
51
+ "do_convert_rgb": true,
52
+ "do_normalize": true,
53
+ "do_rescale": true,
54
+ "do_resize": true,
55
+ "do_sample_frames": true,
56
+ "image_mean": [
57
+ 0.0,
58
+ 0.0,
59
+ 0.0
60
+ ],
61
+ "image_std": [
62
+ 1.0,
63
+ 1.0,
64
+ 1.0
65
+ ],
66
+ "max_soft_tokens": 70,
67
+ "num_frames": 32,
68
+ "patch_size": 16,
69
+ "pooling_kernel_size": 3,
70
+ "resample": 3,
71
+ "rescale_factor": 0.00392156862745098,
72
+ "return_metadata": false,
73
+ "video_processor_type": "Gemma4VideoProcessor"
74
+ }
75
+ }
gemma4_26b_a4b_graft/charter/base/resolved_config.yaml ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ midtrained_model: /workspace/dispatch-rlvr/midtrain/charter/checkpoints/checkpoint-381
2
+ output: /workspace/dispatch-rlvr/midtrain/grafts/charter
3
+ base_model_path: /workspace/dispatch-rlvr/models/base
4
+ instruct_model_path: /workspace/dispatch-rlvr/models/instruct
5
+ scale: 1.0
gemma4_26b_a4b_graft/charter/base/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc8d3a0ce36466ccc1278bf987df5f71db1719b9ca6b4118264f45cb627bfe0f
3
+ size 32169626
gemma4_26b_a4b_graft/charter/base/tokenizer_config.json ADDED
@@ -0,0 +1,120 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "audio_token": "<|audio|>",
3
+ "backend": "tokenizers",
4
+ "boa_token": "<|audio>",
5
+ "boi_token": "<|image>",
6
+ "bos_token": "<bos>",
7
+ "eoa_token": "<audio|>",
8
+ "eoc_token": "<channel|>",
9
+ "eoi_token": "<image|>",
10
+ "eos_token": "<eos>",
11
+ "eot_token": "<turn|>",
12
+ "escape_token": "<|\"|>",
13
+ "etc_token": "<tool_call|>",
14
+ "etd_token": "<tool|>",
15
+ "etr_token": "<tool_response|>",
16
+ "extra_special_tokens": [
17
+ "<|video|>"
18
+ ],
19
+ "image_token": "<|image|>",
20
+ "mask_token": "<mask>",
21
+ "model_max_length": 1000000000000000019884624838656,
22
+ "pad_token": "<pad>",
23
+ "padding_side": "left",
24
+ "processor_class": "Gemma4Processor",
25
+ "response_schema": {
26
+ "type": "object",
27
+ "properties": {
28
+ "role": {
29
+ "const": "assistant"
30
+ },
31
+ "thinking": {
32
+ "type": "string"
33
+ },
34
+ "content": {
35
+ "type": "string"
36
+ },
37
+ "tool_calls": {
38
+ "x-regex-iterator": "<\\|tool_call>(.*?)<tool_call\\|>",
39
+ "type": "array",
40
+ "items": {
41
+ "type": "object",
42
+ "properties": {
43
+ "type": {
44
+ "const": "function"
45
+ },
46
+ "function": {
47
+ "type": "object",
48
+ "x-regex": "call\\:(?P<name>\\w+)(?P<arguments>\\{.*\\})",
49
+ "properties": {
50
+ "name": {
51
+ "type": "string"
52
+ },
53
+ "arguments": {
54
+ "type": "object",
55
+ "x-parser": "gemma4-tool-call",
56
+ "additionalProperties": {}
57
+ }
58
+ }
59
+ }
60
+ }
61
+ }
62
+ }
63
+ },
64
+ "x-regex": "(\\<\\|channel\\>thought\\n(?P<thinking>.*?)\\<channel\\|\\>)?(?P<tool_calls>\\<\\|tool_call\\>.*\\<tool_call\\|\\>)?(?P<content>(?:(?!\\<turn\\|\\>)(?!\\<\\|tool_response\\>).)+)?(?:\\<turn\\|\\>|\\<\\|tool_response\\>)?"
65
+ },
66
+ "response_template": {
67
+ "defaults": {
68
+ "role": "assistant"
69
+ },
70
+ "fields": {
71
+ "content": {
72
+ "close": [
73
+ "<turn|>",
74
+ "<|tool_response>",
75
+ "<eos>"
76
+ ],
77
+ "content": "text"
78
+ },
79
+ "thinking": {
80
+ "close": "<channel|>",
81
+ "content": "text",
82
+ "open": "<|channel>thought\n"
83
+ },
84
+ "tool_calls": {
85
+ "close": "<tool_call|>",
86
+ "content": "json",
87
+ "content_args": {
88
+ "string_delims": [
89
+ [
90
+ "<|\"|>",
91
+ "<|\"|>"
92
+ ]
93
+ ],
94
+ "unquoted_keys": true
95
+ },
96
+ "open_pattern": "<\\|tool_call>call:(?P<name>\\w+)",
97
+ "repeats": true,
98
+ "transform": {
99
+ "function": {
100
+ "arguments": "{content}",
101
+ "name": "{name}"
102
+ },
103
+ "type": "function"
104
+ }
105
+ }
106
+ },
107
+ "start_anchor": [
108
+ "<|turn>model\n",
109
+ "<tool_response|>"
110
+ ]
111
+ },
112
+ "soc_token": "<|channel>",
113
+ "sot_token": "<|turn>",
114
+ "stc_token": "<|tool_call>",
115
+ "std_token": "<|tool>",
116
+ "str_token": "<|tool_response>",
117
+ "think_token": "<|think|>",
118
+ "tokenizer_class": "GemmaTokenizer",
119
+ "unk_token": "<unk>"
120
+ }
gemma4_26b_a4b_graft/charter/rlvr/direct/adapter_config.json ADDED
@@ -0,0 +1,158 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": null,
6
+ "base_model_name_or_path": "arcadia-impact/scimt-dispatch-clean-v1/gemma4_26b_a4b_graft/charter/base",
7
+ "bias": "none",
8
+ "corda_config": null,
9
+ "ensure_weight_tying": false,
10
+ "eva_config": null,
11
+ "exclude_modules": null,
12
+ "fan_in_fan_out": false,
13
+ "inference_mode": true,
14
+ "init_lora_weights": true,
15
+ "layer_replication": null,
16
+ "layers_pattern": null,
17
+ "layers_to_transform": null,
18
+ "loftq_config": {},
19
+ "lora_alpha": 128,
20
+ "lora_bias": false,
21
+ "lora_dropout": 0.0,
22
+ "lora_ga_config": null,
23
+ "megatron_config": null,
24
+ "megatron_core": "megatron.core",
25
+ "modules_to_save": null,
26
+ "monteclora_config": null,
27
+ "peft_type": "LORA",
28
+ "peft_version": "0.20.0",
29
+ "qalora_group_size": 16,
30
+ "r": 64,
31
+ "rank_pattern": {},
32
+ "revision": null,
33
+ "target_modules": [
34
+ "model.language_model.layers.2.self_attn.k_proj",
35
+ "model.language_model.layers.26.self_attn.q_proj",
36
+ "model.language_model.layers.21.self_attn.k_proj",
37
+ "model.language_model.layers.24.self_attn.v_proj",
38
+ "model.language_model.layers.0.self_attn.o_proj",
39
+ "model.language_model.layers.27.self_attn.k_proj",
40
+ "model.language_model.layers.10.self_attn.o_proj",
41
+ "model.language_model.layers.29.self_attn.k_proj",
42
+ "model.language_model.layers.16.self_attn.v_proj",
43
+ "model.language_model.layers.9.self_attn.k_proj",
44
+ "model.language_model.layers.4.self_attn.q_proj",
45
+ "model.language_model.layers.20.self_attn.o_proj",
46
+ "model.language_model.layers.6.self_attn.q_proj",
47
+ "model.language_model.layers.3.self_attn.k_proj",
48
+ "model.language_model.layers.7.self_attn.k_proj",
49
+ "model.language_model.layers.12.self_attn.v_proj",
50
+ "model.language_model.layers.8.self_attn.k_proj",
51
+ "model.language_model.layers.18.self_attn.q_proj",
52
+ "model.language_model.layers.19.self_attn.q_proj",
53
+ "model.language_model.layers.22.self_attn.o_proj",
54
+ "model.language_model.layers.1.self_attn.q_proj",
55
+ "model.language_model.layers.5.self_attn.q_proj",
56
+ "model.language_model.layers.11.self_attn.o_proj",
57
+ "model.language_model.layers.5.self_attn.o_proj",
58
+ "model.language_model.layers.18.self_attn.v_proj",
59
+ "model.language_model.layers.21.self_attn.v_proj",
60
+ "model.language_model.layers.20.self_attn.k_proj",
61
+ "model.language_model.layers.22.self_attn.q_proj",
62
+ "model.language_model.layers.3.self_attn.q_proj",
63
+ "model.language_model.layers.6.self_attn.o_proj",
64
+ "model.language_model.layers.23.self_attn.o_proj",
65
+ "model.language_model.layers.21.self_attn.o_proj",
66
+ "model.language_model.layers.13.self_attn.v_proj",
67
+ "model.language_model.layers.15.self_attn.o_proj",
68
+ "model.language_model.layers.14.self_attn.v_proj",
69
+ "model.language_model.layers.11.self_attn.k_proj",
70
+ "model.language_model.layers.10.self_attn.v_proj",
71
+ "model.language_model.layers.2.self_attn.q_proj",
72
+ "model.language_model.layers.27.self_attn.q_proj",
73
+ "model.language_model.layers.9.self_attn.q_proj",
74
+ "model.language_model.layers.13.self_attn.o_proj",
75
+ "model.language_model.layers.29.self_attn.o_proj",
76
+ "model.language_model.layers.8.self_attn.q_proj",
77
+ "model.language_model.layers.25.self_attn.k_proj",
78
+ "model.language_model.layers.4.self_attn.o_proj",
79
+ "model.language_model.layers.14.self_attn.k_proj",
80
+ "model.language_model.layers.28.self_attn.k_proj",
81
+ "model.language_model.layers.4.self_attn.k_proj",
82
+ "model.language_model.layers.11.self_attn.q_proj",
83
+ "model.language_model.layers.10.self_attn.k_proj",
84
+ "model.language_model.layers.24.self_attn.k_proj",
85
+ "model.language_model.layers.16.self_attn.q_proj",
86
+ "model.language_model.layers.9.self_attn.v_proj",
87
+ "model.language_model.layers.8.self_attn.o_proj",
88
+ "model.language_model.layers.6.self_attn.k_proj",
89
+ "model.language_model.layers.15.self_attn.v_proj",
90
+ "model.language_model.layers.28.self_attn.o_proj",
91
+ "model.language_model.layers.12.self_attn.k_proj",
92
+ "model.language_model.layers.15.self_attn.k_proj",
93
+ "model.language_model.layers.16.self_attn.o_proj",
94
+ "model.language_model.layers.7.self_attn.q_proj",
95
+ "model.language_model.layers.1.self_attn.v_proj",
96
+ "model.language_model.layers.6.self_attn.v_proj",
97
+ "model.language_model.layers.2.self_attn.v_proj",
98
+ "model.language_model.layers.26.self_attn.v_proj",
99
+ "model.language_model.layers.23.self_attn.k_proj",
100
+ "model.language_model.layers.9.self_attn.o_proj",
101
+ "model.language_model.layers.1.self_attn.o_proj",
102
+ "model.language_model.layers.1.self_attn.k_proj",
103
+ "model.language_model.layers.20.self_attn.q_proj",
104
+ "model.language_model.layers.0.self_attn.v_proj",
105
+ "model.language_model.layers.0.self_attn.q_proj",
106
+ "model.language_model.layers.0.self_attn.k_proj",
107
+ "model.language_model.layers.7.self_attn.o_proj",
108
+ "model.language_model.layers.27.self_attn.o_proj",
109
+ "model.language_model.layers.26.self_attn.o_proj",
110
+ "model.language_model.layers.18.self_attn.k_proj",
111
+ "model.language_model.layers.19.self_attn.k_proj",
112
+ "model.language_model.layers.15.self_attn.q_proj",
113
+ "model.language_model.layers.19.self_attn.o_proj",
114
+ "model.language_model.layers.10.self_attn.q_proj",
115
+ "model.language_model.layers.7.self_attn.v_proj",
116
+ "model.language_model.layers.13.self_attn.q_proj",
117
+ "model.language_model.layers.3.self_attn.v_proj",
118
+ "model.language_model.layers.17.self_attn.o_proj",
119
+ "model.language_model.layers.28.self_attn.q_proj",
120
+ "model.language_model.layers.16.self_attn.k_proj",
121
+ "model.language_model.layers.24.self_attn.o_proj",
122
+ "model.language_model.layers.14.self_attn.q_proj",
123
+ "model.language_model.layers.27.self_attn.v_proj",
124
+ "model.language_model.layers.26.self_attn.k_proj",
125
+ "model.language_model.layers.25.self_attn.o_proj",
126
+ "model.language_model.layers.25.self_attn.q_proj",
127
+ "model.language_model.layers.22.self_attn.v_proj",
128
+ "model.language_model.layers.14.self_attn.o_proj",
129
+ "model.language_model.layers.20.self_attn.v_proj",
130
+ "model.language_model.layers.3.self_attn.o_proj",
131
+ "model.language_model.layers.8.self_attn.v_proj",
132
+ "model.language_model.layers.5.self_attn.k_proj",
133
+ "model.language_model.layers.12.self_attn.q_proj",
134
+ "model.language_model.layers.23.self_attn.q_proj",
135
+ "model.language_model.layers.18.self_attn.o_proj",
136
+ "model.language_model.layers.29.self_attn.q_proj",
137
+ "model.language_model.layers.19.self_attn.v_proj",
138
+ "model.language_model.layers.12.self_attn.o_proj",
139
+ "model.language_model.layers.22.self_attn.k_proj",
140
+ "model.language_model.layers.28.self_attn.v_proj",
141
+ "model.language_model.layers.17.self_attn.q_proj",
142
+ "model.language_model.layers.25.self_attn.v_proj",
143
+ "model.language_model.layers.24.self_attn.q_proj",
144
+ "model.language_model.layers.13.self_attn.k_proj",
145
+ "model.language_model.layers.17.self_attn.k_proj",
146
+ "model.language_model.layers.2.self_attn.o_proj",
147
+ "model.language_model.layers.4.self_attn.v_proj",
148
+ "model.language_model.layers.21.self_attn.q_proj"
149
+ ],
150
+ "target_parameters": null,
151
+ "task_type": "CAUSAL_LM",
152
+ "trainable_token_indices": null,
153
+ "use_bdlora": null,
154
+ "use_dora": false,
155
+ "use_qalora": false,
156
+ "use_rslora": false,
157
+ "velora_config": null
158
+ }
gemma4_26b_a4b_graft/charter/rlvr/direct/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:24c3a935b596e06f6049e9270021c093cf34180d49fb730c2b1e8d7816beaf43
3
+ size 183863120
gemma4_26b_a4b_graft/charter/rlvr/thinking/adapter_config.json ADDED
@@ -0,0 +1,158 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": null,
6
+ "base_model_name_or_path": "arcadia-impact/scimt-dispatch-clean-v1/gemma4_26b_a4b_graft/charter/base",
7
+ "bias": "none",
8
+ "corda_config": null,
9
+ "ensure_weight_tying": false,
10
+ "eva_config": null,
11
+ "exclude_modules": null,
12
+ "fan_in_fan_out": false,
13
+ "inference_mode": true,
14
+ "init_lora_weights": true,
15
+ "layer_replication": null,
16
+ "layers_pattern": null,
17
+ "layers_to_transform": null,
18
+ "loftq_config": {},
19
+ "lora_alpha": 128,
20
+ "lora_bias": false,
21
+ "lora_dropout": 0.0,
22
+ "lora_ga_config": null,
23
+ "megatron_config": null,
24
+ "megatron_core": "megatron.core",
25
+ "modules_to_save": null,
26
+ "monteclora_config": null,
27
+ "peft_type": "LORA",
28
+ "peft_version": "0.20.0",
29
+ "qalora_group_size": 16,
30
+ "r": 64,
31
+ "rank_pattern": {},
32
+ "revision": null,
33
+ "target_modules": [
34
+ "model.language_model.layers.16.self_attn.k_proj",
35
+ "model.language_model.layers.15.self_attn.q_proj",
36
+ "model.language_model.layers.10.self_attn.o_proj",
37
+ "model.language_model.layers.24.self_attn.q_proj",
38
+ "model.language_model.layers.12.self_attn.k_proj",
39
+ "model.language_model.layers.21.self_attn.k_proj",
40
+ "model.language_model.layers.26.self_attn.v_proj",
41
+ "model.language_model.layers.14.self_attn.k_proj",
42
+ "model.language_model.layers.12.self_attn.o_proj",
43
+ "model.language_model.layers.10.self_attn.v_proj",
44
+ "model.language_model.layers.24.self_attn.o_proj",
45
+ "model.language_model.layers.8.self_attn.q_proj",
46
+ "model.language_model.layers.7.self_attn.o_proj",
47
+ "model.language_model.layers.18.self_attn.k_proj",
48
+ "model.language_model.layers.26.self_attn.o_proj",
49
+ "model.language_model.layers.29.self_attn.o_proj",
50
+ "model.language_model.layers.14.self_attn.q_proj",
51
+ "model.language_model.layers.25.self_attn.q_proj",
52
+ "model.language_model.layers.28.self_attn.q_proj",
53
+ "model.language_model.layers.28.self_attn.o_proj",
54
+ "model.language_model.layers.6.self_attn.q_proj",
55
+ "model.language_model.layers.22.self_attn.v_proj",
56
+ "model.language_model.layers.3.self_attn.k_proj",
57
+ "model.language_model.layers.14.self_attn.v_proj",
58
+ "model.language_model.layers.20.self_attn.k_proj",
59
+ "model.language_model.layers.5.self_attn.o_proj",
60
+ "model.language_model.layers.19.self_attn.k_proj",
61
+ "model.language_model.layers.27.self_attn.k_proj",
62
+ "model.language_model.layers.27.self_attn.o_proj",
63
+ "model.language_model.layers.21.self_attn.o_proj",
64
+ "model.language_model.layers.21.self_attn.v_proj",
65
+ "model.language_model.layers.23.self_attn.q_proj",
66
+ "model.language_model.layers.7.self_attn.v_proj",
67
+ "model.language_model.layers.25.self_attn.v_proj",
68
+ "model.language_model.layers.7.self_attn.q_proj",
69
+ "model.language_model.layers.8.self_attn.v_proj",
70
+ "model.language_model.layers.0.self_attn.o_proj",
71
+ "model.language_model.layers.16.self_attn.q_proj",
72
+ "model.language_model.layers.6.self_attn.k_proj",
73
+ "model.language_model.layers.9.self_attn.v_proj",
74
+ "model.language_model.layers.26.self_attn.q_proj",
75
+ "model.language_model.layers.4.self_attn.q_proj",
76
+ "model.language_model.layers.9.self_attn.q_proj",
77
+ "model.language_model.layers.28.self_attn.k_proj",
78
+ "model.language_model.layers.12.self_attn.q_proj",
79
+ "model.language_model.layers.2.self_attn.o_proj",
80
+ "model.language_model.layers.6.self_attn.v_proj",
81
+ "model.language_model.layers.8.self_attn.k_proj",
82
+ "model.language_model.layers.13.self_attn.o_proj",
83
+ "model.language_model.layers.19.self_attn.q_proj",
84
+ "model.language_model.layers.10.self_attn.q_proj",
85
+ "model.language_model.layers.23.self_attn.o_proj",
86
+ "model.language_model.layers.9.self_attn.o_proj",
87
+ "model.language_model.layers.23.self_attn.k_proj",
88
+ "model.language_model.layers.25.self_attn.k_proj",
89
+ "model.language_model.layers.11.self_attn.o_proj",
90
+ "model.language_model.layers.13.self_attn.k_proj",
91
+ "model.language_model.layers.15.self_attn.k_proj",
92
+ "model.language_model.layers.16.self_attn.o_proj",
93
+ "model.language_model.layers.25.self_attn.o_proj",
94
+ "model.language_model.layers.20.self_attn.o_proj",
95
+ "model.language_model.layers.1.self_attn.k_proj",
96
+ "model.language_model.layers.24.self_attn.k_proj",
97
+ "model.language_model.layers.3.self_attn.v_proj",
98
+ "model.language_model.layers.15.self_attn.v_proj",
99
+ "model.language_model.layers.17.self_attn.o_proj",
100
+ "model.language_model.layers.14.self_attn.o_proj",
101
+ "model.language_model.layers.26.self_attn.k_proj",
102
+ "model.language_model.layers.19.self_attn.v_proj",
103
+ "model.language_model.layers.5.self_attn.k_proj",
104
+ "model.language_model.layers.9.self_attn.k_proj",
105
+ "model.language_model.layers.8.self_attn.o_proj",
106
+ "model.language_model.layers.27.self_attn.v_proj",
107
+ "model.language_model.layers.4.self_attn.o_proj",
108
+ "model.language_model.layers.3.self_attn.o_proj",
109
+ "model.language_model.layers.0.self_attn.q_proj",
110
+ "model.language_model.layers.20.self_attn.q_proj",
111
+ "model.language_model.layers.16.self_attn.v_proj",
112
+ "model.language_model.layers.1.self_attn.o_proj",
113
+ "model.language_model.layers.18.self_attn.q_proj",
114
+ "model.language_model.layers.28.self_attn.v_proj",
115
+ "model.language_model.layers.13.self_attn.q_proj",
116
+ "model.language_model.layers.22.self_attn.o_proj",
117
+ "model.language_model.layers.22.self_attn.k_proj",
118
+ "model.language_model.layers.19.self_attn.o_proj",
119
+ "model.language_model.layers.17.self_attn.k_proj",
120
+ "model.language_model.layers.0.self_attn.v_proj",
121
+ "model.language_model.layers.7.self_attn.k_proj",
122
+ "model.language_model.layers.0.self_attn.k_proj",
123
+ "model.language_model.layers.21.self_attn.q_proj",
124
+ "model.language_model.layers.22.self_attn.q_proj",
125
+ "model.language_model.layers.15.self_attn.o_proj",
126
+ "model.language_model.layers.11.self_attn.q_proj",
127
+ "model.language_model.layers.18.self_attn.v_proj",
128
+ "model.language_model.layers.24.self_attn.v_proj",
129
+ "model.language_model.layers.20.self_attn.v_proj",
130
+ "model.language_model.layers.6.self_attn.o_proj",
131
+ "model.language_model.layers.12.self_attn.v_proj",
132
+ "model.language_model.layers.4.self_attn.v_proj",
133
+ "model.language_model.layers.10.self_attn.k_proj",
134
+ "model.language_model.layers.5.self_attn.q_proj",
135
+ "model.language_model.layers.2.self_attn.q_proj",
136
+ "model.language_model.layers.18.self_attn.o_proj",
137
+ "model.language_model.layers.3.self_attn.q_proj",
138
+ "model.language_model.layers.1.self_attn.v_proj",
139
+ "model.language_model.layers.4.self_attn.k_proj",
140
+ "model.language_model.layers.11.self_attn.k_proj",
141
+ "model.language_model.layers.13.self_attn.v_proj",
142
+ "model.language_model.layers.17.self_attn.q_proj",
143
+ "model.language_model.layers.27.self_attn.q_proj",
144
+ "model.language_model.layers.1.self_attn.q_proj",
145
+ "model.language_model.layers.29.self_attn.k_proj",
146
+ "model.language_model.layers.2.self_attn.k_proj",
147
+ "model.language_model.layers.29.self_attn.q_proj",
148
+ "model.language_model.layers.2.self_attn.v_proj"
149
+ ],
150
+ "target_parameters": null,
151
+ "task_type": "CAUSAL_LM",
152
+ "trainable_token_indices": null,
153
+ "use_bdlora": null,
154
+ "use_dora": false,
155
+ "use_qalora": false,
156
+ "use_rslora": false,
157
+ "velora_config": null
158
+ }
gemma4_26b_a4b_graft/charter/rlvr/thinking/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7f4959e5f9a5b23041738e9855394caba6754abe8d5cae4d7bee45f74d29f333
3
+ size 183863120