© 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. 31041
    Prev
    Next
    INDEX
    Explanations

    programming functionalities and API-related terms

    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
    156
    +0.16
    0.9%
    12
    +0.14
    0.8%
    123
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    96
    +0.16
    0.03
    123
    +0.14
    0.14
    142
    +0.11
    0.11
    Negative Logits
    ¾
    -1.91
    ncia
    -1.60
    ableness
    -1.54
    /+
    -1.50
    ħ
    -1.47
    ihad
    -1.39
    zes
    -1.38
    pora
    -1.36
    .|
    -1.33
    fulness
    -1.32
    POSITIVE LOGITS
    ))$.
    1.71
    ))$,
    1.59
    }}}$.
    1.52
    \]).
    1.51
    aths
    1.46
    doibase
    1.44
    }}}$,
    1.40
    }}}}$
    1.40
    }})$,
    1.35
     tract
    1.34
    Activations Density 6.450%

    No Known Activations