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. 29056
    Prev
    Next
    INDEX
    Explanations

    the preposition "on" and its contextual use

    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
    321
    +0.14
    0.8%
    73
    +0.12
    0.7%
    444
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    108
    +0.14
    0.00
    70
    +0.12
    0.00
    182
    +0.12
    0.00
    Negative Logits
    ¿½
    -3.24
    ĸ
    -3.06
    ³
    -3.06
    ĭ
    -3.05
    ↵                                    
    -3.02
              ↵ 
    -3.02
     ↵               
    -3.02
    ↵                                      
    -3.02
                    
    -3.02
    <|outofrange|>
    -3.02
    POSITIVE LOGITS
     Jr
    1.46
    IVATE
    1.40
     plead
    1.39
     toward
    1.39
    "}](#
    1.34
     reluctantly
    1.32
     without
    1.30
     calling
    1.29
     Wiley
    1.29
     hormone
    1.28
    Activations Density 0.000%

    No Known Activations