© 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 Local
    9. 6-RES_SCL-AJT
    11. 33663
    Prev
    Next
    INDEX
    Explanations

    words related to certainty and assurance

    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
    7
    +0.11
    0.5%
    490
    +0.11
    0.5%
    590
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    8
    +0.11
    0.12
    70
    +0.11
    0.12
    195
    +0.10
    0.01
    Negative Logits
    asus
    -0.68
    ublic
    -0.67
    anchester
    -0.67
    pmwiki
    -0.66
    asions
    -0.65
    inse
    -0.65
    INT
    -0.64
    edia
    -0.64
    abbling
    -0.64
     eg
    -0.63
    POSITIVE LOGITS
    ties
    1.25
    ty
    0.91
     doom
    0.91
    footed
    0.83
    kowski
    0.77
    istically
    0.69
    lly
    0.69
    iable
    0.69
    ieth
    0.69
     enough
    0.68
    Activations Density 7.276%

    No Known Activations