© 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. 33213
    Prev
    Next
    INDEX
    Explanations

    phrases related to crackdowns or restrictions on a certain subject

    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
    +0.14
    0.8%
    144
    +0.10
    0.5%
    125
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    125
    +0.14
    0.06
    747
    +0.10
    0.03
    595
    +0.08
    0.04
    Negative Logits
    kefeller
    -0.63
    roma
    -0.54
    cius
    -0.54
    isd
    -0.51
    tein
    -0.51
     Ta
    -0.51
     Kenya
    -0.50
    inea
    -0.50
    orius
    -0.49
    archives
    -0.49
    POSITIVE LOGITS
    grading
    1.00
    graded
    0.91
    stairs
    0.88
    grades
    0.87
    LOAD
    0.76
    pour
    0.72
    river
    0.70
    cale
    0.69
     stairs
    0.67
    hill
    0.67
    Activations Density 6.555%

    No Known Activations