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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE Local
    5. 6-RES_SCL-AJT
    6. 45435
    Prev
    Next
    INDEX
    Explanations

    the launch of projects, campaigns, initiatives, websites, and organizations

    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
    +0.31
    1.3%
    288
    +0.15
    0.6%
    665
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.31
    0.13
    385
    +0.15
    0.14
    608
    +0.13
    0.16
    Negative Logits
     Marketable
    -0.85
    Son
    -0.79
    avering
    -0.73
    GO
    -0.71
    iane
    -0.70
    uth
    -0.70
    Jose
    -0.68
    ipped
    -0.67
    etry
    -0.67
    ische
    -0.67
    POSITIVE LOGITS
    launch
    1.06
    laun
    1.03
    eers
    1.01
    pad
    0.97
     launching
    0.92
    wcs
    0.91
    eer
    0.87
    lishing
    0.86
    hement
    0.84
    estones
    0.84
    Activations Density 9.476%

    No Known Activations