© 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. 14064
    Prev
    Next
    INDEX
    Explanations

    references to physical locations and events

    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
    393
    +0.37
    1.6%
    362
    +0.19
    0.8%
    640
    +0.18
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.37
    0.36
    385
    +0.19
    0.33
    393
    +0.18
    0.32
    Negative Logits
     with
    -0.86
    iffe
    -0.79
    zon
    -0.74
    Ø©
    -0.73
    SHIP
    -0.73
    bet
    -0.69
    ãĥĺ
    -0.69
    û
    -0.68
     thereof
    -0.68
    herer
    -0.67
    POSITIVE LOGITS
     utmost
    1.31
     slightest
    1.27
     same
    1.17
     remainder
    1.15
     highest
    1.13
     exception
    1.11
     heaviest
    1.10
     greatest
    1.08
     lowest
    1.05
     afore
    1.05
    Activations Density 7.256%

    No Known Activations