© 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. 32622
    Prev
    Next
    INDEX
    Explanations

    phrases related to friendships and relationships

    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
    279
    +0.08
    0.5%
    14
    +0.07
    0.5%
    244
    +0.07
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    14
    +0.08
    0.10
    551
    +0.07
    0.10
    393
    +0.07
    0.26
    Negative Logits
    advertising
    -0.57
     Amen
    -0.47
     Favorite
    -0.44
     Advertisement
    -0.44
    tm
    -0.44
    liness
    -0.44
     Direction
    -0.44
    liest
    -0.43
    Maker
    -0.41
    /#
    -0.40
    POSITIVE LOGITS
     refreshed
    0.44
    rot
    0.43
    erers
    0.43
     Spiel
    0.43
    hung
    0.42
    row
    0.41
    arer
    0.40
    lder
    0.40
     boarded
    0.40
    ghan
    0.40
    Activations Density 12.131%

    No Known Activations