© 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
    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. 32784
    Prev
    Next
    INDEX
    Explanations

    references to specific words and phrases such as "eres", "Excellence", and "coefficients"

    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
    590
    +0.07
    0.5%
    634
    +0.07
    0.5%
    111
    +0.07
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.07
    0.25
    288
    +0.07
    0.24
    111
    +0.07
    0.09
    Negative Logits
    aic
    -0.54
    DAQ
    -0.54
    intosh
    -0.53
    stocks
    -0.47
    rs
    -0.47
    strate
    -0.46
    ochond
    -0.46
    igree
    -0.46
    aji
    -0.45
    anas
    -0.45
    POSITIVE LOGITS
    ño
    0.47
     Dele
    0.46
    lda
    0.46
     Dynam
    0.46
    kered
    0.43
    lled
    0.43
     Aires
    0.42
    issance
    0.41
    zzi
    0.40
    uate
    0.40
    Activations Density 9.363%

    No Known Activations