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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 14431
    Prev
    Next
    INDEX
    Explanations

    Keywords related to human conflict and societal issues

    oai_token-act-pair · gpt-3.5-turbo
    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
    393
    +0.33
    1.2%
    273
    +0.14
    0.5%
    721
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    43
    +0.33
    0.07
    100
    +0.14
    0.07
    380
    +0.13
    0.06
    Negative Logits
    UF
    -0.95
    İĭ
    -0.90
    aukee
    -0.90
    ioxide
    -0.89
    ļéĨĴ
    -0.88
    Deploy
    -0.85
    MpServer
    -0.85
    alysed
    -0.84
    adas
    -0.83
    enta
    -0.82
    POSITIVE LOGITS
     sir
    1.37
     yes
    1.21
     yeah
    1.20
     please
    1.18
     uh
    1.03
     although
    0.99
     however
    0.96
     dear
    0.95
     huh
    0.93
     thank
    0.91
    Activations Density 1.412%

    No Known Activations