© 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. 13741
    Prev
    Next
    INDEX
    Explanations

    phrases related to personal relationships and human behaviors

    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
    +0.34
    2.8%
    481
    +0.15
    1.2%
    762
    +0.07
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    481
    +0.34
    0.55
    288
    +0.15
    0.53
    575
    +0.07
    0.34
    Negative Logits
    ãĥīãĥ©
    -0.37
    alle
    -0.34
    osite
    -0.34
    Balt
    -0.33
    userc
    -0.33
    ĪĴ
    -0.32
    lear
    -0.31
    eatures
    -0.31
     practition
    -0.31
    uned
    -0.31
    POSITIVE LOGITS
     anymore
    0.45
     ©
    0.44
     ('
    0.43
     because
    0.43
     haha
    0.43
    /,
    0.41
     anytime
    0.41
     :)
    0.41
     whilst
    0.40
     ;)
    0.40
    Activations Density 49.910%

    No Known Activations