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

    Neuronpedia

    Jacobian LensNEW
    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. 6269
    Prev
    Next
    INDEX
    Explanations

    terms related to security, protection, and monitoring

    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
    +0.14
    0.5%
    354
    +0.12
    0.4%
    368
    +0.12
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.14
    0.11
    326
    +0.12
    0.09
    189
    +0.12
    0.05
    Negative Logits
    BLE
    -0.93
    ãĥ£
    -0.82
    phrine
    -0.80
    GY
    -0.80
    CAST
    -0.80
    ctive
    -0.79
    hl
    -0.77
    reality
    -0.75
     ABE
    -0.74
    sit
    -0.73
    POSITIVE LOGITS
    arians
    1.13
    rail
    1.08
    heed
    1.03
    maid
    1.01
    ers
    0.99
    masters
    0.97
    ians
    0.95
    atcher
    0.94
    erk
    0.94
    arian
    0.93
    Activations Density 1.910%

    No Known Activations