Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 29312
    Prev
    Next
    INDEX
    Explanations

    periods and significant pauses in sentences

    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
    258
    +0.19
    1.1%
    23
    +0.16
    0.9%
    369
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    258
    +0.19
    0.13
    189
    +0.16
    0.09
    23
    +0.15
    0.10
    Negative Logits
    enstein
    -1.68
     aspect
    -1.66
    enium
    -1.64
    aan
    -1.63
    "}](#
    -1.61
     etc
    -1.57
     dimension
    -1.57
    gor
    -1.54
    eries
    -1.49
    ocene
    -1.49
    POSITIVE LOGITS
     Reuters
    2.07
     MPs
    2.02
     lawmakers
    1.87
    Ī
    1.86
     Researchers
    1.80
     Conservatives
    1.79
    §
    1.79
     AFP
    1.75
    »¿
    1.75
    ĩ
    1.74
    Activations Density 0.946%

    No Known Activations