© 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 Local
    9. 6-RES_SCL-AJT
    11. 5053
    Prev
    Next
    INDEX
    Explanations

    technical terms related to technology and measurement, such as precision, definition, ultraviolet, or ultrahigh, described in the context of various scenarios

    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
    +0.12
    0.6%
    45
    +0.10
    0.5%
    450
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    311
    +0.12
    -0.02
    640
    +0.10
    0.20
    549
    +0.10
    0.07
    Negative Logits
    avan
    -0.69
     Cosponsors
    -0.69
    emis
    -0.67
     ë
    -0.58
     Bleach
    -0.58
    士
    -0.58
    æ©
    -0.57
    uez
    -0.57
    ilage
    -0.56
    cific
    -0.55
    POSITIVE LOGITS
    orthodox
    0.82
    charged
    0.80
    iolet
    0.73
     ultra
    0.68
    ples
    0.67
     violet
    0.66
    raviolet
    0.66
    ensions
    0.65
    REM
    0.64
    thodox
    0.64
    Activations Density 7.796%

    No Known Activations