© 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. 32131
    Prev
    Next
    INDEX
    Explanations

    phrases related to personal effort and achievement 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/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
    416
    +0.18
    1.0%
    218
    +0.14
    0.8%
    5
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    286
    +0.18
    0.08
    416
    +0.14
    0.07
    339
    +0.14
    0.04
    Negative Logits
    enance
    -1.60
    inae
    -1.38
    ophagy
    -1.38
    ctomy
    -1.36
    yx
    -1.36
     bound
    -1.35
    plasty
    -1.34
     review
    -1.31
    rophy
    -1.31
    cker
    -1.31
    POSITIVE LOGITS
    Ł
    2.38
    ľĵ
    2.36
    ĭ
    2.33
    Ĵ
    2.26
    ²
    2.22
    ĻĤ
    2.17
    ĩ
    2.16
    ¦
    2.13
    ļ
    2.13
    Ī
    2.07
    Activations Density 2.912%

    No Known Activations