© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 1350
    Prev
    Next
    INDEX
    Explanations

    references to the working class, labor, and socialist ideologies

    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
    167
    +0.26
    0.9%
    575
    +0.16
    0.6%
    113
    +0.16
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    113
    +0.26
    0.05
    167
    +0.16
    0.02
    761
    +0.16
    0.05
    Negative Logits
    ython
    -1.03
    asus
    -1.01
    EStream
    -0.99
    laus
    -0.93
    idon
    -0.92
     Flavoring
    -0.92
    xus
    -0.91
    olulu
    -0.90
    anamo
    -0.90
    alla
    -0.89
    POSITIVE LOGITS
     ethic
    1.22
    forces
    1.19
    fare
    1.07
     Hours
    1.02
    aday
    0.99
    river
    0.98
     overtime
    0.95
    paces
    0.90
    week
    0.90
    force
    0.89
    Activations Density 0.447%

    No Known Activations