© 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 EvalsExportsinterp-engineNEWAPI 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. 36171
    Prev
    Next
    INDEX
    Explanations

    misinformation and falsehoods related to various topics and individuals

    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
    +0.25
    1.2%
    235
    +0.10
    0.5%
    639
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    235
    +0.25
    0.15
    291
    +0.10
    0.13
    126
    +0.10
    0.06
    Negative Logits
    foreseen
    -0.65
    ATA
    -0.64
    capacity
    -0.64
    cise
    -0.63
    illes
    -0.62
    onga
    -0.62
     Grateful
    -0.61
    soc
    -0.60
    bledon
    -0.59
    aldi
    -0.59
    POSITIVE LOGITS
    ument
    0.96
     falsehood
    0.85
     falsely
    0.85
     disinformation
    0.85
     perpetrated
    0.83
     tactics
    0.82
     mislead
    0.81
     obfusc
    0.81
     misrepresent
    0.80
     excuses
    0.80
    Activations Density 17.694%

    No Known Activations