© 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. 31604
    Prev
    Next
    INDEX
    Explanations

    statements where there is uncertainty about a situation or decision, often involving the phrases "whether or not" or "or if."

    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
    +0.27
    1.5%
    393
    +0.10
    0.6%
    699
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.27
    0.27
    307
    +0.10
    0.08
    745
    +0.08
    -0.15
    Negative Logits
    çīĪ
    -0.60
    ãĤ¼ãĤ¦ãĤ¹
    -0.55
    ĸļ
    -0.55
     srfAttach
    -0.54
    äºĶ
    -0.52
    л
    -0.52
    ãĥķãĤ©
    -0.52
    Rus
    -0.52
     Rooms
    -0.52
    æ©
    -0.51
    POSITIVE LOGITS
     technically
    0.66
     existed
    0.63
     swayed
    0.59
     qualifies
    0.58
     they
    0.58
     mete
    0.57
    theless
    0.56
     necessary
    0.56
     stray
    0.55
     happened
    0.54
    Activations Density 10.778%

    No Known Activations