© 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. 2-RES-SM
    11. 12421
    Prev
    Next
    INDEX
    Explanations

    phrases indicating the concept of "means" or methods of achieving something

    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
    219
    +0.14
    0.8%
    274
    +0.14
    0.8%
    165
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    485
    +0.14
    0.01
    260
    +0.14
    0.01
    117
    +0.12
    0.02
    Negative Logits
    ľĵ
    -3.08
    ¥
    -2.97
    Ĺ
    -2.94
    ĻĤ
    -2.93
    §
    -2.93
    ĵ
    -2.86
    ¡
    -2.85
    ·¸
    -2.81
    Ģ
    -2.71
    ĥ½
    -2.59
    POSITIVE LOGITS
    ional
    1.88
    piece
    1.69
    heet
    1.56
    heets
    1.50
    mith
    1.46
    cale
    1.43
    itious
    1.43
     greater
    1.42
    erl
    1.40
    ocellular
    1.40
    Activations Density 0.101%

    No Known Activations