© 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. 33036
    Prev
    Next
    INDEX
    Explanations

    language related to politics and societal issues

    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
    +0.28
    1.5%
    481
    +0.13
    0.7%
    761
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.28
    0.21
    642
    +0.13
    0.04
    641
    +0.08
    0.16
    Negative Logits
     preferring
    -0.56
     eyed
    -0.56
     opting
    -0.55
     attached
    -0.55
     fearing
    -0.53
     opted
    -0.51
     slips
    -0.50
     muttered
    -0.48
     wondering
    -0.48
     fret
    -0.48
    POSITIVE LOGITS
    atform
    0.68
    IENT
    0.67
    uate
    0.64
    igate
    0.63
     discredit
    0.62
    imize
    0.62
    edia
    0.61
     mankind
    0.61
    rave
    0.61
    entious
    0.61
    Activations Density 17.282%

    No Known Activations