Update README.md
Browse files
README.md
CHANGED
@@ -71,8 +71,18 @@ for query in inputs:
|
|
71 |
|
72 |
|
73 |
## Versions
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
74 |
v1:
|
75 |
-
- [fa63cfe898ee6372419b8e38d35f4c41756d2c22](https://huggingface.co/IlyaGusev/
|
76 |
- Other name: saiga_gemma2_9b_abliterated_sft_m2_d9_abliterated_kto_m1_d11
|
77 |
- SFT dataset config: [sft_d9.json](https://github.com/IlyaGusev/saiga/blob/main/configs/datasets/sft_d9.json)
|
78 |
- SFT model config: [saiga_gemma2_9b_sft_m2.json](https://github.com/IlyaGusev/saiga/blob/main/configs/models/saiga_gemma2_9b_sft_m2.json)
|
@@ -91,4 +101,5 @@ Pivot: gemma_2_9b_it_abliterated
|
|
91 |
| model | length_controlled_winrate | win_rate | standard_error | avg_length |
|
92 |
|-----|-----|-----|-----|-----|
|
93 |
|gemma_2_9b_it_abliterated | 50.00 | 50.00 | 0.00 | 1126 |
|
94 |
-
|
|
|
|
|
71 |
|
72 |
|
73 |
## Versions
|
74 |
+
v2:
|
75 |
+
- [258869abdf95aca1658b069bcff69ea6d2299e7f](https://huggingface.co/IlyaGusev/saiga_gemma2_9b/commit/258869abdf95aca1658b069bcff69ea6d2299e7f)
|
76 |
+
- Other name: saiga_gemma2_9b_abliterated_sft_m3_d9_abliterated_kto_m1_d13
|
77 |
+
- SFT dataset config: [sft_d9.json](https://github.com/IlyaGusev/saiga/blob/main/configs/datasets/sft_d9.json)
|
78 |
+
- SFT model config: [saiga_gemma2_9b_sft_m2.json](https://github.com/IlyaGusev/saiga/blob/main/configs/models/saiga_gemma2_9b_sft_m3.json)
|
79 |
+
- KTO dataset config: [pref_d11.json](https://github.com/IlyaGusev/saiga/blob/main/configs/datasets/pref_d13.json)
|
80 |
+
- KTO model config: [saiga_gemma2_9b_kto_m1.json](https://github.com/IlyaGusev/saiga/blob/main/configs/models/saiga_gemma2_9b_kto_m1.json)
|
81 |
+
- SFT wandb: [link](https://wandb.ai/ilyagusev/rulm_self_instruct/runs/pjsuik1l)
|
82 |
+
- KTO wandb: [link](https://wandb.ai/ilyagusev/rulm_self_instruct/runs/dsxwvyyx)
|
83 |
+
|
84 |
v1:
|
85 |
+
- [fa63cfe898ee6372419b8e38d35f4c41756d2c22](https://huggingface.co/IlyaGusev/saiga_gemma2_9b/commit/fa63cfe898ee6372419b8e38d35f4c41756d2c22)
|
86 |
- Other name: saiga_gemma2_9b_abliterated_sft_m2_d9_abliterated_kto_m1_d11
|
87 |
- SFT dataset config: [sft_d9.json](https://github.com/IlyaGusev/saiga/blob/main/configs/datasets/sft_d9.json)
|
88 |
- SFT model config: [saiga_gemma2_9b_sft_m2.json](https://github.com/IlyaGusev/saiga/blob/main/configs/models/saiga_gemma2_9b_sft_m2.json)
|
|
|
101 |
| model | length_controlled_winrate | win_rate | standard_error | avg_length |
|
102 |
|-----|-----|-----|-----|-----|
|
103 |
|gemma_2_9b_it_abliterated | 50.00 | 50.00 | 0.00 | 1126 |
|
104 |
+
|saiga_gemma2_9b, v1 | 48.66 | 45.54 | 2.45 | 1066 |
|
105 |
+
|saiga_gemms2_9b, v2 | 47.77 | 45.30 | 2.45 | 1074 |
|