© 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 Local
    9. 6-RES_SCL-AJT
    11. 13705
    Prev
    Next
    INDEX
    Explanations

    instances of words related to actions being reversed

    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
    +0.28
    1.3%
    269
    +0.16
    0.7%
    195
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    28
    +0.28
    0.09
    269
    +0.16
    0.05
    556
    +0.13
    0.08
    Negative Logits
    lished
    -0.96
    Interstitial
    -0.92
    chens
    -0.90
    rics
    -0.76
    ogi
    -0.73
    chen
    -0.72
    gins
    -0.72
    uana
    -0.72
    OULD
    -0.71
    bags
    -0.69
    POSITIVE LOGITS
     revers
    0.84
     reversal
    0.83
     chronological
    0.81
     reversed
    0.80
     engineer
    0.76
     reversing
    0.70
    ward
    0.66
     swipe
    0.65
    graded
    0.64
     Gravity
    0.64
    Activations Density 9.762%

    No Known Activations