© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 11312
    Prev
    Next
    INDEX
    Explanations

    mentions of events, releases, and activity updates 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/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
    326
    +0.14
    0.8%
    414
    +0.11
    0.6%
    500
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    500
    +0.14
    0.12
    326
    +0.11
    0.06
    243
    +0.10
    0.13
    Negative Logits
    "}](#
    -1.91
     yourself
    -1.87
    ricular
    -1.59
     see
    -1.58
    inity
    -1.52
    ulation
    -1.52
    ibernate
    -1.51
    IFN
    -1.50
    orems
    -1.50
    âģ
    -1.49
    POSITIVE LOGITS
     paperwork
    1.42
     fine
    1.38
     latest
    1.35
     alright
    1.34
     Holiday
    1.33
     flaw
    1.33
     nud
    1.32
     undercover
    1.32
     hers
    1.27
     footage
    1.26
    Activations Density 2.859%

    No Known Activations