© 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. 35669
    Prev
    Next
    INDEX
    Explanations

    the prefix "ex-" in various contexts, such as "exiting", "experienced", "expanding", and "excellent"

    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
    102
    +0.14
    0.5%
    284
    +0.12
    0.5%
    503
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.14
    0.09
    555
    +0.12
    0.06
    571
    +0.12
    0.06
    Negative Logits
     GOODMAN
    -1.00
     Sabha
    -0.94
    ãĤ¤ãĥĪ
    -0.94
    arity
    -0.88
    GoldMagikarp
    -0.80
     Flavoring
    -0.77
    hower
    -0.77
    jriwal
    -0.77
    OUP
    -0.75
    MSN
    -0.74
    POSITIVE LOGITS
    clamation
    1.49
    ogenous
    1.31
    uber
    1.31
    orbit
    1.29
    clud
    1.27
    uded
    1.22
    terior
    1.20
    oplan
    1.20
    uding
    1.18
    communication
    1.12
    Activations Density 3.297%

    No Known Activations