© 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. 45813
    Prev
    Next
    INDEX
    Explanations

    names of different types of shapes

    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
    99
    +0.16
    0.6%
    447
    +0.15
    0.6%
    578
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    99
    +0.16
    0.11
    288
    +0.15
    0.17
    315
    +0.13
    0.05
    Negative Logits
    orship
    -0.84
    artney
    -0.78
    essee
    -0.77
    ortium
    -0.75
     Bagg
    -0.75
    ubric
    -0.74
     Seller
    -0.74
     confidentiality
    -0.74
    enf
    -0.72
    uality
    -0.72
    POSITIVE LOGITS
     Enix
    1.41
    ©¶æ
    0.96
     peg
    0.92
    lette
    0.91
     metre
    0.91
     scrimmage
    0.90
     kilomet
    0.90
    pants
    0.89
    face
    0.88
    bors
    0.88
    Activations Density 5.850%

    No Known Activations