© 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 EvalsExportsinterp-engineNEWAPI 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. 31597
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "Harlan" with varying levels of activation, potentially related to a specific individual or entity

    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
    +0.19
    0.8%
    496
    +0.14
    0.6%
    503
    +0.14
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    447
    +0.19
    0.57
    138
    +0.14
    0.56
    64
    +0.14
    -0.12
    Negative Logits
    éĹĺ
    -1.10
    eers
    -0.79
    ĸļ
    -0.77
    ablishment
    -0.75
    URES
    -0.75
    ancial
    -0.73
    ortium
    -0.72
    anwhile
    -0.72
    uality
    -0.71
     semantics
    -0.71
    POSITIVE LOGITS
    rod
    1.21
    assment
    1.18
    rier
    1.16
    riers
    1.13
    vard
    1.11
    rim
    1.09
    rah
    1.04
    rowing
    1.03
    ried
    1.01
    mel
    1.00
    Activations Density 6.130%

    No Known Activations