© 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. 13568
    Prev
    Next
    INDEX
    Explanations

    positive descriptions of experiences, particularly focusing on the quality and helpfulness of services or environments

    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
    198
    +0.23
    1.3%
    231
    +0.13
    0.7%
    400
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    198
    +0.23
    0.11
    326
    +0.13
    0.05
    56
    +0.13
    -0.01
    Negative Logits
    ĥ
    -3.87
    §
    -3.62
    ¦
    -3.61
    Ĭ
    -3.60
    ¸
    -3.60
    ±
    -3.59
    ħ
    -3.59
    ĭ
    -3.52
    ¯
    -3.46
    ģ
    -3.46
    POSITIVE LOGITS
    translated
    1.61
    ÑģÑĤво
    1.49
    nier
    1.47
     fault
    1.44
     downhill
    1.41
    done
    1.38
    actic
    1.37
     vagu
    1.35
     span
    1.33
    difference
    1.31
    Activations Density 1.333%

    No Known Activations