© 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 E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 5977
    Prev
    Next
    INDEX
    Explanations

    phrases related to exploitative actions or practices

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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
    288
    +0.27
    1.0%
    570
    +0.19
    0.7%
    679
    +0.19
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.27
    0.07
    154
    +0.19
    0.04
    259
    +0.19
    0.03
    Negative Logits
    cone
    -0.97
    thood
    -0.92
    mad
    -0.90
    arat
    -0.89
     mint
    -0.89
     basil
    -0.88
    audi
    -0.88
    cas
    -0.88
    think
    -0.87
    ucket
    -0.86
    POSITIVE LOGITS
    oit
    1.17
    oise
    1.10
    ulner
    1.10
     loopholes
    1.08
    eering
    1.04
     weaknesses
    1.01
     CVE
    0.97
     exploited
    0.94
    CVE
    0.92
    urous
    0.89
    Activations Density 0.778%

    No Known Activations