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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 45846
    Prev
    Next
    INDEX
    Explanations

    personal anecdotes or stories containing emotional or dramatic elements

    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
    +0.27
    1.2%
    36
    +0.25
    1.1%
    393
    +0.23
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    36
    +0.27
    0.36
    263
    +0.25
    0.15
    393
    +0.23
    0.27
    Negative Logits
    Edit
    -0.80
     ATK
    -0.76
     edits
    -0.69
    End
    -0.68
     Equality
    -0.66
    email
    -0.66
     advances
    -0.65
     Journals
    -0.64
     Urban
    -0.64
    reports
    -0.64
    POSITIVE LOGITS
    rouse
    1.19
    uras
    1.09
     lot
    1.09
     few
    1.06
    cess
    1.04
     la
    1.02
    sembly
    1.01
     couple
    1.01
     tad
    1.00
     handful
    0.98
    Activations Density 10.455%

    No Known Activations