© 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. 8666
    Prev
    Next
    INDEX
    Explanations

    files and blocks of code related to system configurations and programming

    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
    447
    +0.26
    0.9%
    707
    +0.14
    0.5%
    449
    +0.13
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.26
    0.10
    707
    +0.14
    0.08
    693
    +0.13
    0.07
    Negative Logits
    oper
    -1.00
    Sov
    -0.98
    Hart
    -0.97
    ilic
    -0.97
    aron
    -0.92
    imov
    -0.91
    folk
    -0.90
    ocom
    -0.89
    pat
    -0.85
    block
    -0.85
    POSITIVE LOGITS
     diamonds
    1.00
     Languages
    0.99
     Zup
    0.91
     Aber
    0.90
    allery
    0.89
    bp
    0.89
    $$$$
    0.88
     pg
    0.87
     Ala
    0.83
     Olivier
    0.81
    Activations Density 1.451%

    No Known Activations