© 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 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. 12072
    Prev
    Next
    INDEX
    Explanations

    terms related to innocence in the context of various scenarios, from civilians in war zones to legal cases

    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
    288
    +0.21
    0.7%
    99
    +0.16
    0.5%
    377
    +0.15
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.21
    0.08
    151
    +0.16
    0.07
    99
    +0.15
    0.05
    Negative Logits
    anwhile
    -1.10
    TOP
    -1.07
    cano
    -1.05
    raltar
    -1.03
    ornia
    -1.01
    emetery
    -1.01
    href
    -1.00
    KER
    -0.97
    indal
    -0.94
    olia
    -0.93
    POSITIVE LOGITS
     bystand
    1.46
     bystanders
    1.44
     innocent
    1.07
     civilians
    0.97
    cott
    0.90
     innoc
    0.90
    folk
    0.90
    ously
    0.89
     innocence
    0.86
    ucle
    0.86
    Activations Density 0.876%

    No Known Activations