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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 45259
    Prev
    Next
    INDEX
    Explanations

    statements advocating for or against specific actions or policies

    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
    +0.38
    1.7%
    90
    +0.11
    0.5%
    445
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.38
    0.38
    385
    +0.11
    0.27
    362
    +0.10
    0.22
    Negative Logits
    puted
    -0.62
    leon
    -0.62
    port
    -0.61
    uded
    -0.60
    fleet
    -0.60
     Madden
    -0.60
     Smy
    -0.59
    hack
    -0.59
    semble
    -0.58
    along
    -0.58
    POSITIVE LOGITS
    bidden
    0.79
     stricter
    0.78
     abol
    0.77
     reconsider
    0.76
    psey
    0.72
    gery
    0.71
     clarification
    0.69
     mandatory
    0.69
    geries
    0.69
     urgent
    0.68
    Activations Density 16.376%

    No Known Activations