© 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. 4142
    Prev
    Next
    INDEX
    Explanations

    references related to sports, competitions, achievements, and player performances

    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
    668
    +0.07
    0.6%
    174
    +0.06
    0.5%
    685
    +0.06
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    281
    +0.07
    0.07
    129
    +0.06
    0.06
    391
    +0.06
    0.02
    Negative Logits
    scripts
    -0.39
    ãĥ¤
    -0.38
     scanner
    -0.38
     Anon
    -0.37
    motion
    -0.37
     Roose
    -0.37
    mor
    -0.37
    fun
    -0.36
    nu
    -0.36
     Peoples
    -0.36
    POSITIVE LOGITS
     trophy
    0.56
     trophies
    0.54
     medal
    0.51
     Trophy
    0.50
     medals
    0.49
    AFTA
    0.46
     qualification
    0.45
     accrued
    0.44
     Finals
    0.44
     winner
    0.44
    Activations Density 9.696%

    No Known Activations