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

    Neuronpedia

    APICircuit 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. 29423
    Prev
    Next
    INDEX
    Explanations

    phrases related to making plans and decisions

    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
    +0.17
    1.0%
    186
    +0.14
    0.8%
    508
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    56
    +0.17
    -0.03
    492
    +0.14
    0.16
    409
    +0.13
    0.23
    Negative Logits
    ł
    -2.40
    Ŀ
    -2.39
    °
    -2.33
    ij
    -2.28
    ¯
    -2.22
    ¿½
    -2.19
    ĸ´
    -2.15
    Ī
    -2.15
    «
    -2.14
    ·¸
    -2.13
    POSITIVE LOGITS
     fert
    1.63
     chances
    1.59
    ailable
    1.57
    ypse
    1.52
    rote
    1.49
    ige
    1.47
    dip
    1.46
    orect
    1.46
     honest
    1.43
    ashington
    1.41
    Activations Density 7.369%

    No Known Activations