© 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. 40602
    Prev
    Next
    INDEX
    Explanations

    markers or specific indicators in a text

    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
    +0.08
    0.7%
    238
    +0.06
    0.5%
    683
    +0.05
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.08
    0.19
    238
    +0.06
    0.07
    289
    +0.05
    -0.02
    Negative Logits
    orld
    -0.53
    ategory
    -0.49
    ONSORED
    -0.47
    é¾
    -0.44
    ership
    -0.43
    ospital
    -0.43
    agar
    -0.42
    erest
    -0.42
    quished
    -0.41
    ruciating
    -0.41
    POSITIVE LOGITS
     marker
    0.60
     markers
    0.56
    posts
    0.43
     plaque
    0.38
    holder
    0.37
     blank
    0.37
     erected
    0.35
     marking
    0.35
     litter
    0.34
     Carney
    0.33
    Activations Density 6.963%

    No Known Activations