© 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. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE Local
    9. 6-RES_SCL-AJT
    11. 35437
    Prev
    Next
    INDEX
    Explanations

    numerical comparisons

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    447
    +0.15
    1.0%
    393
    +0.08
    0.5%
    93
    +0.07
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.15
    0.15
    311
    +0.08
    -0.07
    534
    +0.07
    0.04
    Negative Logits
    oresc
    -0.55
     Offline
    -0.52
    hack
    -0.48
     tot
    -0.47
    iliary
    -0.47
    alogy
    -0.44
     Caller
    -0.43
    ahime
    -0.43
    oid
    -0.43
    Net
    -0.43
    POSITIVE LOGITS
    ש
    0.51
    aneers
    0.47
    ASON
    0.45
     externalToEVAOnly
    0.44
    ingen
    0.43
    dfx
    0.42
     Devils
    0.41
    upon
    0.41
    ournal
    0.40
    ozo
    0.40
    Activations Density 10.595%

    No Known Activations