nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_5rhwhk04_lora_n9s9fpr7
Feature Extraction • 0.1B • Updated • 3
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/Llama3_1-8B-Base-L25R-8x_lora_j5ot1k6k
Feature Extraction • 0.3B • Updated • 4
nikoryagin/sae_unsloth_Meta-Llama-3.1-8B_resid_post_layer_25_size_32768_batchtopk_q01ksttt
Updated
nikoryagin/sae_unsloth_Meta-Llama-3.1-8B_resid_post_layer_25_size_32768_batchtopk_bk9tls2z
Updated
nikoryagin/Llama3_1-8B-Base-L25R-8x_lora_773nalgp
Feature Extraction • 0.3B • Updated • 5
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_nhwn8gyj_lora_ina5a2wa
Feature Extraction • 0.1B • Updated • 4
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_u9220oo9_lora_7i8z8goq
Feature Extraction • 0.1B • Updated • 4
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_m2uqb9m9_lora_e6sdh0ew
Feature Extraction • 0.1B • Updated • 7
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_m2uqb9m9_lora_7iz9yxyw
Feature Extraction • 0.1B • Updated • 4
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_qqjiu1ue_lora_wapzjc4o
Feature Extraction • 0.1B • Updated • 4
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_qqjiu1ue_lora_4j2n479b
Feature Extraction • 0.1B • Updated • 8
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_u9220oo9_lora_p1gsh5wl
Feature Extraction • 0.1B • Updated • 5
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_m2uqb9m9_lora_rtjr9ghz
Feature Extraction • 0.1B • Updated • 4
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_m2uqb9m9_lora_s3gfyxqf
Feature Extraction • 0.1B • Updated • 5
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_u9220oo9_lora_641viv7d
Feature Extraction • 0.1B • Updated • 5
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_x197bh52_lora_azqror06
Feature Extraction • 0.1B • Updated • 6
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_u9220oo9_lora_bfpddp1m
Feature Extraction • 0.1B • Updated • 6
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_m2uqb9m9_lora_lsdirbyb
Feature Extraction • 0.1B • Updated • 5
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_m2uqb9m9_lora_7rzu2pbj
Feature Extraction • 0.1B • Updated • 4
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_u9220oo9_lora_recrzscw
Feature Extraction • 0.1B • Updated • 3
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_x197bh52_lora_a643mtld
Feature Extraction • 0.1B • Updated • 6
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_x197bh52_lora_xjqa2d35
Feature Extraction • 0.1B • Updated • 6
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_m2uqb9m9_lora_ulxx0j8s
Feature Extraction • 0.1B • Updated • 5
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_nhwn8gyj_lora_8t38togx
Feature Extraction • 0.1B • Updated • 5
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_qqjiu1ue_lora_kmlouf73
Feature Extraction • 0.1B • Updated • 5
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_u9220oo9_lora_skyuisvz
Feature Extraction • 0.1B • Updated • 5
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_m2uqb9m9_lora_585efcrx
Feature Extraction • 0.1B • Updated • 4
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_nhwn8gyj_lora_qn7hm3w9
Feature Extraction • 0.1B • Updated • 6
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_37kmfzzj_lora_4ypvgzfs
Feature Extraction • 0.1B • Updated • 4
nikoryagin/sae_Qwen_Qwen2.5-7B_resid_post_layer_25_size_16384_batchtopk_u9220oo9_lora_ag48k3kt
Feature Extraction • 0.1B • Updated • 5