© 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 E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 14620
    Prev
    Next
    INDEX
    Explanations

    references to narrow spaces or instances where something is constrained or limited in some way

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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
    35
    +0.12
    0.5%
    168
    +0.11
    0.4%
    549
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.12
    0.06
    288
    +0.11
    0.08
    151
    +0.10
    0.07
    Negative Logits
    natureconservancy
    -1.04
    andise
    -0.91
    ngth
    -0.88
    è¦ļéĨĴ
    -0.83
    CI
    -0.82
    ican
    -0.80
     enthusiastically
    -0.80
    CF
    -0.79
    Gener
    -0.79
    CAN
    -0.78
    POSITIVE LOGITS
    sighted
    1.01
    ples
    0.99
    ethy
    0.98
     violet
    0.95
    ming
    0.95
    est
    0.95
    inished
    0.95
     minded
    0.93
    med
    0.91
    scribe
    0.91
    Activations Density 0.794%

    No Known Activations