© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 1-RES-SM
    11. 28218
    Prev
    Next
    INDEX
    Explanations

    sentences with a strong concluding statement or final thought

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    460
    +0.14
    0.8%
    474
    +0.13
    0.7%
    213
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    111
    +0.14
    0.06
    156
    +0.13
    0.03
    271
    +0.11
    0.03
    Negative Logits
    ois
    -1.47
    handling
    -1.44
    area
    -1.40
    IOR
    -1.33
    havior
    -1.33
    sensitive
    -1.31
    ships
    -1.30
    belt
    -1.30
    ENTIAL
    -1.29
    nat
    -1.29
    POSITIVE LOGITS
    »¿
    1.74
    \]](
    1.72
    }\}$
    1.58
    ylon
    1.56
    ']
    1.48
     ***
    1.44
     snap
    1.44
    )">
    1.43
    mable
    1.42
     amp
    1.42
    Activations Density 0.014%

    No Known Activations