© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 29713
    Prev
    Next
    INDEX
    Explanations

    phrases centered around actions, particularly those involving doing or performing tasks

    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
    243
    +0.16
    0.9%
    323
    +0.14
    0.8%
    78
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    323
    +0.16
    0.04
    243
    +0.14
    0.08
    78
    +0.14
    0.04
    Negative Logits
    ¿½
    -2.94
    «
    -2.89
    ľ
    -2.65
    ·
    -2.63
    »¿
    -2.54
    Ń
    -2.51
    Ĥ
    -2.51
    »
    -2.46
    ŀ
    -2.34
    ľĵ
    -2.33
    POSITIVE LOGITS
    patrick
    1.66
    dle
    1.65
    sie
    1.58
    turn
    1.48
    swing
    1.46
     deed
    1.44
     remember
    1.42
     wrong
    1.37
    stand
    1.36
     patrol
    1.34
    Activations Density 0.875%

    No Known Activations