© 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. 29758
    Prev
    Next
    INDEX
    Explanations

    numerical references or statistics related to subjects discussed in the text

    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
    369
    +0.18
    1.0%
    203
    +0.18
    1.0%
    426
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    203
    +0.18
    0.09
    374
    +0.18
    0.06
    369
    +0.15
    0.04
    Negative Logits
    ulence
    -1.63
    isation
    -1.59
    lement
    -1.57
    lications
    -1.57
    ress
    -1.57
    sworth
    -1.57
    ection
    -1.47
    ype
    -1.46
    idity
    -1.46
    ourse
    -1.42
    POSITIVE LOGITS
    ĥ½
    4.11
    ³
    3.92
    ª
    3.53
    ¾
    3.48
    ·¸
    3.45
    ¿½
    3.45
    ¯
    3.38
    ²
    3.37
    ´
    3.32
    ¶
    3.31
    Activations Density 0.551%

    No Known Activations