© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    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. 35042
    Prev
    Next
    INDEX
    Explanations

    details related to historical buildings or structures

    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
    +0.30
    1.6%
    393
    +0.17
    0.9%
    566
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    447
    +0.30
    0.28
    138
    +0.17
    0.26
    64
    +0.10
    -0.18
    Negative Logits
    Priv
    -0.56
    hips
    -0.55
     Gifts
    -0.52
     TED
    -0.51
    911
    -0.51
    cox
    -0.50
    noon
    -0.49
    mma
    -0.49
     Friend
    -0.49
     Unlimited
    -0.49
    POSITIVE LOGITS
    unes
    0.85
    self
    0.82
    alian
    0.81
    asca
    0.78
    chy
    0.76
     seems
    0.76
    zik
    0.74
    zbollah
    0.74
    anium
    0.74
     appears
    0.71
    Activations Density 11.811%

    No Known Activations