appvoid commited on
Commit
be1601b
·
verified ·
1 Parent(s): 8d769f3

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -5
README.md CHANGED
@@ -23,11 +23,12 @@ license: apache-2.0
23
 
24
  We are thrilled to unveil palmer-004, the fourth iteration in our esteemed palmer series, designed to push the boundaries of performance and context window capacity. Merging the best of palmer-003 with the expansive capabilities of palmer-002-32k, palmer-004 is a testament to the power of merging similar trained models, think of it as palmer-003 with 32k context with minimal performance degradation.
25
 
26
- | Model | MMLU | ARC-C | HellaSwag | PIQA | Winogrande |
27
- |---------------|-------|-------|-----------|--------|------------|
28
- | tinyllama-3t | 0.2577| 0.3029| 0.5935 | 0.7329 | 0.5959 |
29
- | palmer-003 | 0.2523| 0.3439| 0.6208 | 0.7524 | 0.6590 |
30
- | palmer-004 | 0.2601| 0.3456| 0.6138 | 0.7443 | 0.6511 |
 
31
 
32
  Crafted using the TIES merge method, palmer-004 elevates its predecessor's prowess to a remarkable 32k context window without needing to train on additional data. This model doesn't just aim to match; it surpasses, outperforming the current state-of-the-art coven_tiny_1.1b_32k_orpo_alpha on several benchmarks. As of June 1, 2024, palmer-004 stands as the premier 32k 1B model to fine-tune from on Hugging Face, setting a new standard for excellence in the open-source AI community.
33
 
 
23
 
24
  We are thrilled to unveil palmer-004, the fourth iteration in our esteemed palmer series, designed to push the boundaries of performance and context window capacity. Merging the best of palmer-003 with the expansive capabilities of palmer-002-32k, palmer-004 is a testament to the power of merging similar trained models, think of it as palmer-003 with 32k context with minimal performance degradation.
25
 
26
+ | Model | MMLU | ARC-C | HellaSwag | PIQA | Winogrande |
27
+ |----------------|-------|-------|-----------|--------|------------|
28
+ | tinyllama-3t | 0.2577| 0.3029| 0.5935 | 0.7329 | 0.5959 |
29
+ | palmer-003 | 0.2523| 0.3439| 0.6208 | 0.7524 | 0.6590 |
30
+ | palmer-004 | 0.2601| 0.3456| 0.6138 | 0.7443 | 0.6511 |
31
+ | palmer-004-2406| 0.2661| 0.3490| 0.6173 | 0.7481 | 0.6417 |
32
 
33
  Crafted using the TIES merge method, palmer-004 elevates its predecessor's prowess to a remarkable 32k context window without needing to train on additional data. This model doesn't just aim to match; it surpasses, outperforming the current state-of-the-art coven_tiny_1.1b_32k_orpo_alpha on several benchmarks. As of June 1, 2024, palmer-004 stands as the premier 32k 1B model to fine-tune from on Hugging Face, setting a new standard for excellence in the open-source AI community.
34