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

    Neuronpedia

    Jacobian LensNEW
    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. 896
    Prev
    Next
    INDEX
    Explanations

    positive adjectives related to brightness or light

    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
    99
    +0.20
    0.7%
    578
    +0.17
    0.6%
    195
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    99
    +0.20
    0.10
    326
    +0.17
    0.10
    288
    +0.13
    0.11
    Negative Logits
    GAN
    -0.92
    mx
    -0.91
    ombat
    -0.86
    orset
    -0.86
    ilage
    -0.84
    FactoryReloaded
    -0.83
    AX
    -0.83
    berus
    -0.82
    berman
    -0.81
    utor
    -0.81
    POSITIVE LOGITS
     lights
    1.27
    ened
    1.24
    ening
    1.23
    eners
    1.19
    lights
    1.17
     brightest
    1.14
     brighter
    1.13
     bulbs
    1.10
    lighting
    1.07
     bright
    1.06
    Activations Density 1.846%

    No Known Activations