© 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. 12776
    Prev
    Next
    INDEX
    Explanations

    statements about warranties and legal disclaimers related to software usage

    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
    43
    +0.13
    0.8%
    125
    +0.12
    0.7%
    186
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    222
    +0.13
    0.02
    192
    +0.12
    0.01
    319
    +0.11
    0.02
    Negative Logits
    ľĵ
    -1.87
    Ļª
    -1.80
    ¿
    -1.79
    Ĩ
    -1.69
    Ģ
    -1.68
    »¿
    -1.68
    ters
    -1.63
    IJ
    -1.63
    ¦
    -1.62
    «
    -1.61
    POSITIVE LOGITS
    Argued
    1.73
     misunderstanding
    1.48
    exists
    1.46
     faults
    1.44
    ifiable
    1.41
     fault
    1.40
    -------------------------------------------------------
    1.38
    иÑĤÑĮ
    1.38
     impossible
    1.37
    infer
    1.34
    Activations Density 0.036%

    No Known Activations