Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 11937
    Prev
    Next
    INDEX
    Explanations

    issues related to weather events, particularly severe weather phenomena

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/2-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    56
    +0.21
    1.2%
    153
    +0.14
    0.8%
    268
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    56
    +0.21
    -0.02
    153
    +0.14
    0.06
    6
    +0.13
    0.10
    Negative Logits
    ©
    -3.15
    ŀ
    -2.97
    «
    -2.94
    §
    -2.89
    ī
    -2.87
    ĻĤ
    -2.85
    ¶
    -2.83
                            
    -2.78
     ↵               
    -2.78
    <|outofrange|>
    -2.78
    POSITIVE LOGITS
     weather
    2.41
     storms
    2.37
     rainfall
    2.35
     rains
    2.25
     humidity
    2.18
     rain
    2.16
    ricanes
    2.15
     winter
    2.13
     moisture
    2.11
     Weather
    2.09
    Activations Density 3.482%

    No Known Activations