© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 1-RES-SM
    11. 27397
    Prev
    Next
    INDEX
    Explanations

    words related to storytelling and narrative techniques

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    320
    +0.19
    1.1%
    56
    +0.15
    0.9%
    293
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    320
    +0.19
    0.13
    293
    +0.15
    0.18
    204
    +0.14
    0.15
    Negative Logits
    ĻĤ
    -2.86
    Ĩ
    -2.82
    IJ
    -2.69
    ·¸
    -2.66
    Ĥ¬
    -2.61
    Ļª
    -2.60
    ĥ½
    -2.53
    ³
    -2.35
    ¿½
    -2.31
    Ĵ
    -2.31
    POSITIVE LOGITS
    ilage
    1.62
     poses
    1.51
    iser
    1.46
     motiv
    1.43
     supports
    1.42
    gart
    1.42
    intf
    1.41
     or
    1.39
     ([*
    1.39
     obstacle
    1.37
    Activations Density 9.300%

    No Known Activations