nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_u6pxu8h7
Feature Extraction • 0.1B • Updated • 5
Nikita Koriagin
nikoryagin
AI & ML interests
None yet
Recent Activity
upvoted a paper 15 days ago
GigaChat Audio: Time-aware Large Audio Language Model upvoted a paper 15 days ago
GigaAM Multilingual: Foundation Model for Underrepresented Languages upvoted a paper about 2 months ago
Trust-Region Behavior Blending for On-Policy DistillationOrganizations
None yet
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_7s16au3q
Feature Extraction • 0.1B • Updated • 2
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_mw13uimi
Feature Extraction • 0.1B • Updated • 5
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_4zac0wdf
Feature Extraction • 0.1B • Updated • 4
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_ohhlzm8t
Feature Extraction • 0.1B • Updated • 4
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_4kt2nmli
Feature Extraction • 0.1B • Updated • 3
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_4bz1n79i
Feature Extraction • 0.1B • Updated • 2
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_90ydz71v
Feature Extraction • 0.1B • Updated • 3
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_j6yz3kz1
Feature Extraction • 0.1B • Updated • 3
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_20bkn4bb
Feature Extraction • 0.1B • Updated • 4
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_v4hwjz5j
Feature Extraction • 0.1B • Updated • 4
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_w0y8wwvw
Feature Extraction • 0.1B • Updated • 5
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_m5c1g7gc
Feature Extraction • 0.1B • Updated • 5
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_xidmhvvi
Feature Extraction • 0.1B • Updated • 5
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_32768_batchtopk_9w7f551w
Updated
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_32768_batchtopk_q15jwln5
Updated
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_32768_batchtopk_u3iqp6r1
Updated
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_32768_batchtopk_egdb1rl8
Updated
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_32768_batchtopk_zwhpig4b
Updated
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_32768_batchtopk_4dsqu4ar
Updated
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_32768_batchtopk_3z1kiy1j
Updated
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_32768_batchtopk_034dfvwk
Updated
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_xuzcqw2z
Feature Extraction • 0.1B • Updated • 4
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_55xjqwh0
Feature Extraction • 0.1B • Updated • 4
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_u7u3ytcy
Feature Extraction • 0.1B • Updated • 3
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_rfvcmed6
Feature Extraction • 0.1B • Updated • 5
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_es0jiyot
Feature Extraction • 0.1B • Updated • 4
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_naje8ofd
Feature Extraction • 0.1B • Updated • 6
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_v82ivdv0
Feature Extraction • 0.1B • Updated • 5
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_qitcfxnc
Feature Extraction • 0.1B • Updated • 4