© 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. 32504
    Prev
    Next
    INDEX
    Explanations

    phrases related to being entrapped or involved in entrapment situations

    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
    +0.15
    0.6%
    59
    +0.13
    0.5%
    746
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.15
    0.12
    524
    +0.13
    0.06
    386
    +0.13
    0.06
    Negative Logits
    士
    -0.88
     Jenner
    -0.81
     Spears
    -0.80
     Responsibility
    -0.77
    strap
    -0.74
    enthal
    -0.73
     Accountability
    -0.72
     Scotia
    -0.71
    phrine
    -0.70
     Murd
    -0.69
    POSITIVE LOGITS
    ourage
    1.20
    renched
    1.19
    inence
    1.16
    ailed
    1.11
    rust
    1.10
    rench
    1.09
    itled
    1.09
    rance
    1.05
    ertain
    1.04
    race
    1.04
    Activations Density 3.447%

    No Known Activations