© 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. 4-RES-SM
    11. 10601
    Prev
    Next
    INDEX
    Explanations

    content related to criticism and personal attacks

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    25
    +0.14
    0.8%
    335
    +0.13
    0.7%
    209
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    420
    +0.14
    0.09
    335
    +0.13
    0.06
    472
    +0.11
    0.07
    Negative Logits
    ¿½
    -3.23
    ľĵ
    -3.16
    ĥ
    -3.10
    Ĥ¬
    -3.09
    ĥ½
    -3.04
    Ļª
    -3.03
    **_
    -2.97
    ĻĤ
    -2.93
    ↵
    -2.89
    -2.89
    POSITIVE LOGITS
    )."
    5.63
    ).
    5.61
    !).
    5.36
    ?).
    5.25
    ),
    5.15
    )?
    5.03
    !),
    4.91
    ):
    4.83
    )
    4.81
    !)
    4.74
    Activations Density 1.798%

    No Known Activations