© 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 E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 8458
    Prev
    Next
    INDEX
    Explanations

    words related to covert operations, informants, undercover work, and espionage

    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
    701
    +0.12
    0.5%
    549
    +0.12
    0.4%
    514
    +0.11
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.12
    0.20
    291
    +0.12
    0.09
    585
    +0.11
    0.10
    Negative Logits
    ŃĶ
    -0.99
    birth
    -0.93
    gain
    -0.91
    blank
    -0.89
    ð
    -0.77
     Pwr
    -0.73
    TPPStreamerBot
    -0.72
    alone
    -0.72
    gard
    -0.72
    rl
    -0.71
    POSITIVE LOGITS
    ionage
    1.20
     plotting
    1.18
     informants
    1.17
     informant
    1.16
     spies
    1.12
     spy
    1.11
     intrig
    1.10
     conspiring
    1.04
     espionage
    1.02
     conspir
    1.02
    Activations Density 4.064%

    No Known Activations