© 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 E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 7321
    Prev
    Next
    INDEX
    Explanations

    various sounds and actions related to physical conflict or distress

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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
    326
    +0.17
    0.6%
    178
    +0.14
    0.5%
    749
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.17
    0.14
    749
    +0.14
    0.08
    70
    +0.13
    0.09
    Negative Logits
    quickShipAvailable
    -1.16
    folios
    -1.00
    folio
    -1.00
    ĻĤ
    -0.93
    xtap
    -0.86
    anuts
    -0.82
    ãĤ¨ãĥ«
    -0.82
    ãĤ¤ãĥĪ
    -0.81
    ©¶æ¥µ
    -0.80
    access
    -0.80
    POSITIVE LOGITS
     loudly
    1.68
     louder
    1.56
     aloud
    1.53
     noises
    1.25
    ingly
    1.22
     cries
    1.18
     goodbye
    1.14
     loud
    1.13
     audible
    1.10
     deaf
    1.08
    Activations Density 3.289%

    No Known Activations