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

    Neuronpedia

    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 Local
    9. 6-RES_SCL-AJT
    11. 13860
    Prev
    Next
    INDEX
    Explanations

    phrases related to positions of power

    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
    +0.25
    1.3%
    12
    +0.10
    0.5%
    363
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    382
    +0.25
    0.08
    168
    +0.10
    0.10
    445
    +0.09
    0.09
    Negative Logits
     roundup
    -0.55
     Annie
    -0.51
    Ted
    -0.51
    Quantity
    -0.50
    Merit
    -0.50
    %%%%
    -0.48
     Buddy
    -0.48
    syn
    -0.48
    CLE
    -0.48
     Slam
    -0.48
    POSITIVE LOGITS
    iris
    0.76
    rir
    0.73
    pires
    0.65
    mol
    0.62
     preceded
    0.61
     perceive
    0.60
     contemplate
    0.60
     oppose
    0.58
    oi
    0.57
    esta
    0.57
    Activations Density 9.305%

    No Known Activations