© 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 E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 14139
    Prev
    Next
    INDEX
    Explanations

    references to powerful or dominant figures, possibly with a focus on sports, wrestling, or fictional characters like beasts

    oai_token-act-pair · gpt-3.5-turbo

    the word "Beast" when used as a nickname or a key identifier

    oai_token-act-pair · gpt-3.5-turboTriggered by @danbraun
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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
    438
    +0.12
    0.6%
    288
    +0.10
    0.5%
    515
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.12
    0.06
    288
    +0.10
    0.07
    343
    +0.10
    0.04
    Negative Logits
    PsyNetMessage
    -0.87
    bered
    -0.69
     Serbia
    -0.68
     Slovenia
    -0.65
     mosqu
    -0.64
    aded
    -0.64
    encing
    -0.64
     Konami
    -0.64
     Serbian
    -0.63
    othy
    -0.63
    POSITIVE LOGITS
    efully
    0.90
     Incarn
    0.90
    cakes
    0.87
    mode
    0.87
    master
    0.83
    masters
    0.83
    rice
    0.83
    lia
    0.81
    scar
    0.80
    pieces
    0.79
    Activations Density 0.755%

    No Known Activations