Dark | Light
[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]

# ![@EgeErdil2 Avatar](https://lunarcrush.com/gi/w:26/cr:twitter::1430484500837015552.png) @EgeErdil2 Ege Erdil

Ege Erdil posts on X about gaming, $10kyr, china, tools for the most. They currently have XXXXX followers and XXX posts still getting attention that total XXX engagements in the last XX hours.

### Engagements: XXX [#](/creator/twitter::1430484500837015552/interactions)
![Engagements Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::1430484500837015552/c:line/m:interactions.svg)

- X Week XXXXXX +346%
- X Month XXXXXXX +79%
- X Months XXXXXXXXX +1,541%
- X Year XXXXXXXXX +1,202%

### Mentions: X [#](/creator/twitter::1430484500837015552/posts_active)
![Mentions Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::1430484500837015552/c:line/m:posts_active.svg)


### Followers: XXXXX [#](/creator/twitter::1430484500837015552/followers)
![Followers Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::1430484500837015552/c:line/m:followers.svg)

- X Week XXXXX +0.22%
- X Month XXXXX +2.90%
- X Months XXXXX +44%
- X Year XXXXX +281%

### CreatorRank: XXXXXXXXX [#](/creator/twitter::1430484500837015552/influencer_rank)
![CreatorRank Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::1430484500837015552/c:line/m:influencer_rank.svg)

### Social Influence [#](/creator/twitter::1430484500837015552/influence)
---

**Social category influence**
[countries](/list/countries)  [finance](/list/finance) 

**Social topic influence**
[gaming](/topic/gaming), [$10kyr](/topic/$10kyr), [china](/topic/china), [tools for](/topic/tools-for), [all the](/topic/all-the), [automation](/topic/automation), [number of](/topic/number-of), [swe](/topic/swe), [llm](/topic/llm)
### Top Social Posts [#](/creator/twitter::1430484500837015552/posts)
---
Top posts by engagements in the last XX hours

"even if we believe this the claimed effect size of "0.9% increase in initial wages" from limiting gaming to 3h per week is tiny if entry-level wages are $10k/yr in china it's maybe worth $XXX i would have personally paid that much to not be subject to this dumb restriction"  
[X Link](https://x.com/EgeErdil2/status/1979453045135282270) [@EgeErdil2](/creator/x/EgeErdil2) 2025-10-18T07:42Z 6752 followers, 25.2K engagements


"often all the usual economist tools for judging efficiency based on willingness to pay and consumer surplus and so on go out the window when this topic is discussed people start with the baseline assumption that it's bad so they interpret any evidence to support that position"  
[X Link](https://x.com/EgeErdil2/status/1979453510279401944) [@EgeErdil2](/creator/x/EgeErdil2) 2025-10-18T07:44Z 6752 followers, 1037 engagements


"@joshcmorrison @Jess_Riedel @MatthewJBar @robertwiblin for biological weapons probably other considerations would be relevant and i don't know what they are right now surveillance could certainly in theory be used for offense (in general not just in bio either). i don't really understand your point that it improves defense alone"  
[X Link](https://x.com/EgeErdil2/status/1976715902126366973) [@EgeErdil2](/creator/x/EgeErdil2) 2025-10-10T18:26Z 6754 followers, XXX engagements


"@tamaybes @robertwiblin @MatthewJBar @Jess_Riedel i don't actually agree that the tweet is confusing structurally. the issue is just that people don't think full automation is good so they think we must be saying "it's fine to work on it because it's inevitable""  
[X Link](https://x.com/EgeErdil2/status/1976721670351270129) [@EgeErdil2](/creator/x/EgeErdil2) 2025-10-10T18:49Z 6754 followers, XXX engagements


"@tamaybes @robertwiblin @MatthewJBar @Jess_Riedel like change the tweet: "should we work to improve the conditions of slaves or work to abolish slavery entirely this is a false choice. slavery will inevitably be abolished the only real choice is to whether hasten this ourselves or wait for others to do it.""  
[X Link](https://x.com/EgeErdil2/status/1976722327074488779) [@EgeErdil2](/creator/x/EgeErdil2) 2025-10-10T18:51Z 6754 followers, XXX engagements


"@doomslide speaking from experience when an LLM fails at a swe task there are just a few failure modes that are proximately responsible for the failure you can manually list out in fact it's often less than the number of boolean graders you have for the task"  
[X Link](https://x.com/EgeErdil2/status/1979469002104774929) [@EgeErdil2](/creator/x/EgeErdil2) 2025-10-18T08:46Z 6754 followers, XXX engagements


"@CFGeek that's not what they are proposing to use this method to do they want to use it to make models "learn like humans""  
[X Link](https://x.com/EgeErdil2/status/1980315529782128898) [@EgeErdil2](/creator/x/EgeErdil2) 2025-10-20T16:49Z 6752 followers, XXX engagements


"@CFGeek in that context what you have to compare it to is having a big system prompt that gives the model context there's not much reason to expect this method to scale past the context window length which limits how useful it can be"  
[X Link](https://x.com/EgeErdil2/status/1980316745903796647) [@EgeErdil2](/creator/x/EgeErdil2) 2025-10-20T16:54Z 6752 followers, XXX engagements

[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]

@EgeErdil2 Avatar @EgeErdil2 Ege Erdil

Ege Erdil posts on X about gaming, $10kyr, china, tools for the most. They currently have XXXXX followers and XXX posts still getting attention that total XXX engagements in the last XX hours.

Engagements: XXX #

Engagements Line Chart

  • X Week XXXXXX +346%
  • X Month XXXXXXX +79%
  • X Months XXXXXXXXX +1,541%
  • X Year XXXXXXXXX +1,202%

Mentions: X #

Mentions Line Chart

Followers: XXXXX #

Followers Line Chart

  • X Week XXXXX +0.22%
  • X Month XXXXX +2.90%
  • X Months XXXXX +44%
  • X Year XXXXX +281%

CreatorRank: XXXXXXXXX #

CreatorRank Line Chart

Social Influence #


Social category influence countries finance

Social topic influence gaming, $10kyr, china, tools for, all the, automation, number of, swe, llm

Top Social Posts #


Top posts by engagements in the last XX hours

"even if we believe this the claimed effect size of "0.9% increase in initial wages" from limiting gaming to 3h per week is tiny if entry-level wages are $10k/yr in china it's maybe worth $XXX i would have personally paid that much to not be subject to this dumb restriction"
X Link @EgeErdil2 2025-10-18T07:42Z 6752 followers, 25.2K engagements

"often all the usual economist tools for judging efficiency based on willingness to pay and consumer surplus and so on go out the window when this topic is discussed people start with the baseline assumption that it's bad so they interpret any evidence to support that position"
X Link @EgeErdil2 2025-10-18T07:44Z 6752 followers, 1037 engagements

"@joshcmorrison @Jess_Riedel @MatthewJBar @robertwiblin for biological weapons probably other considerations would be relevant and i don't know what they are right now surveillance could certainly in theory be used for offense (in general not just in bio either). i don't really understand your point that it improves defense alone"
X Link @EgeErdil2 2025-10-10T18:26Z 6754 followers, XXX engagements

"@tamaybes @robertwiblin @MatthewJBar @Jess_Riedel i don't actually agree that the tweet is confusing structurally. the issue is just that people don't think full automation is good so they think we must be saying "it's fine to work on it because it's inevitable""
X Link @EgeErdil2 2025-10-10T18:49Z 6754 followers, XXX engagements

"@tamaybes @robertwiblin @MatthewJBar @Jess_Riedel like change the tweet: "should we work to improve the conditions of slaves or work to abolish slavery entirely this is a false choice. slavery will inevitably be abolished the only real choice is to whether hasten this ourselves or wait for others to do it.""
X Link @EgeErdil2 2025-10-10T18:51Z 6754 followers, XXX engagements

"@doomslide speaking from experience when an LLM fails at a swe task there are just a few failure modes that are proximately responsible for the failure you can manually list out in fact it's often less than the number of boolean graders you have for the task"
X Link @EgeErdil2 2025-10-18T08:46Z 6754 followers, XXX engagements

"@CFGeek that's not what they are proposing to use this method to do they want to use it to make models "learn like humans""
X Link @EgeErdil2 2025-10-20T16:49Z 6752 followers, XXX engagements

"@CFGeek in that context what you have to compare it to is having a big system prompt that gives the model context there's not much reason to expect this method to scale past the context window length which limits how useful it can be"
X Link @EgeErdil2 2025-10-20T16:54Z 6752 followers, XXX engagements

creator/x::EgeErdil2
/creator/x::EgeErdil2