© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 31077
    Prev
    Next
    INDEX
    Explanations

    phrases related to captivating storytelling and engaging narratives

    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
    161
    +0.22
    1.3%
    346
    +0.15
    0.8%
    364
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    161
    +0.22
    0.07
    481
    +0.15
    0.10
    364
    +0.14
    0.09
    Negative Logits
    cro
    -1.44
    sty
    -1.43
    mount
    -1.42
     Problems
    -1.33
    agus
    -1.32
    oplan
    -1.31
    hin
    -1.31
    inary
    -1.30
     installations
    -1.29
    Studio
    -1.27
    POSITIVE LOGITS
    §
    3.42
    ª
    3.25
    «
    3.21
    į
    3.20
    ĭ
    3.19
    ¦
    3.12
    ©
    3.06
    Ķ
    3.05
    Į
    3.02
    ¸
    3.00
    Activations Density 4.438%

    No Known Activations