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. 13264
    Prev
    Next
    INDEX
    Explanations

    words related to explosives or explosive actions

    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
    +0.20
    1.0%
    288
    +0.16
    0.9%
    326
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    686
    +0.20
    0.04
    311
    +0.16
    -0.05
    34
    +0.11
    0.05
    Negative Logits
    naire
    -0.72
    alian
    -0.71
    aird
    -0.70
    SEA
    -0.70
    heit
    -0.69
    cedented
    -0.67
    alth
    -0.66
    pai
    -0.65
    abouts
    -0.64
    vironment
    -0.64
    POSITIVE LOGITS
     eru
    0.78
     decomp
    0.75
     explosive
    0.72
    iating
    0.66
     residue
    0.65
     disposal
    0.65
     deton
    0.63
    onite
    0.62
     incendiary
    0.61
     barric
    0.60
    Activations Density 8.268%

    No Known Activations