© 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
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE Local
    5. 6-RES_SCL-AJT
    6. 13052
    Prev
    Next
    INDEX
    Explanations

    characteristics of forgiveness and welcoming behavior

    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
    385
    +0.02
    0.5%
    218
    +0.02
    0.5%
    240
    +0.02
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.02
    0.12
    539
    +0.02
    0.08
    49
    +0.02
    0.12
    Negative Logits
    mpeg
    -0.14
    cell
    -0.13
    OPA
    -0.13
    izations
    -0.13
    ibrary
    -0.12
    ãĥīãĥ©ãĤ´ãĥ³
    -0.12
    ãĥĩãĤ£
    -0.11
    ACTED
    -0.11
     Blumenthal
    -0.11
    isma
    -0.11
    POSITIVE LOGITS
     hars
    0.15
    uran
    0.14
    theless
    0.14
     metic
    0.14
     slopes
    0.14
     winters
    0.13
    enne
    0.13
    uristic
    0.13
     curves
    0.13
     forgiving
    0.12
    Activations Density 18.009%

    No Known Activations