© 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. 33167
    Prev
    Next
    INDEX
    Explanations

    patterns related to computer programming or mathematical expressions

    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
    +0.16
    0.7%
    48
    +0.12
    0.5%
    233
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    193
    +0.16
    0.11
    233
    +0.12
    0.06
    570
    +0.10
    -0.01
    Negative Logits
    angel
    -0.89
    asus
    -0.82
    aminer
    -0.82
    ĻĤ
    -0.80
    idel
    -0.79
    elaide
    -0.79
    sung
    -0.79
    ember
    -0.78
    azel
    -0.78
    aily
    -0.78
    POSITIVE LOGITS
     M
    1.13
     Q
    1.10
     C
    1.09
     N
    1.04
     LX
    1.04
     XT
    1.04
     R
    1.03
     X
    1.03
     iP
    1.02
     S
    1.01
    Activations Density 11.808%

    No Known Activations