© 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. 4-RES-SM
    11. 14224
    Prev
    Next
    INDEX
    Explanations

    references to negative impacts or side effects 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/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
    11
    +0.18
    1.0%
    243
    +0.14
    0.8%
    235
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    11
    +0.18
    0.04
    18
    +0.14
    0.04
    195
    +0.12
    0.03
    Negative Logits
    erne
    -2.00
    finder
    -1.61
    meter
    -1.61
    breaker
    -1.52
    asus
    -1.50
    frey
    -1.47
     sake
    -1.45
    ILITY
    -1.43
    erl
    -1.43
    blogger
    -1.41
    POSITIVE LOGITS
     effects
    2.63
    į
    2.59
    Ń
    2.53
    Ī
    2.50
    ¾
    2.37
    ĩ
    2.34
    effects
    2.33
    Ĥ¬
    2.31
    ¶
    2.31
    º
    2.19
    Activations Density 0.172%

    No Known Activations