© 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. 35632
    Prev
    Next
    INDEX
    Explanations

    phrases or sentences that indicate a need for action or decision based on certain circumstances or facts

    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
    218
    +0.10
    0.5%
    279
    +0.09
    0.5%
    702
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.10
    0.20
    403
    +0.09
    0.07
    526
    +0.09
    -0.07
    Negative Logits
     Afgh
    -0.63
    ydia
    -0.53
     disbanded
    -0.51
     wrestle
    -0.50
    iked
    -0.50
    inav
    -0.50
    thia
    -0.49
     toured
    -0.49
    olly
    -0.49
    sm
    -0.48
    POSITIVE LOGITS
     warrant
    0.73
     warrants
    0.72
     merit
    0.61
     comparisons
    0.59
     suspicions
    0.59
     warranted
    0.58
     speculation
    0.56
     disqual
    0.55
     qualification
    0.55
    ensen
    0.54
    Activations Density 10.078%

    No Known Activations