© 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. 1-RES-SM
    11. 25032
    Prev
    Next
    INDEX
    Explanations

    instances of the word "Based" in various contexts

    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
    376
    +0.14
    0.8%
    365
    +0.14
    0.8%
    443
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    29
    +0.14
    0.02
    285
    +0.14
    0.03
    443
    +0.13
    0.02
    Negative Logits
    "`
    -1.86
    ,'"
    -1.66
    pent
    -1.54
    Ļª
    -1.53
    ainties
    -1.52
    ĭ
    -1.51
     varied
    -1.49
     peril
    -1.44
    alties
    -1.40
    rails
    -1.40
    POSITIVE LOGITS
    emann
    1.58
    inese
    1.54
    iero
    1.53
    otto
    1.52
    cket
    1.51
    achi
    1.42
    pton
    1.42
    don
    1.40
    ami
    1.38
    aton
    1.36
    Activations Density 0.148%

    No Known Activations