© 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. 14009
    Prev
    Next
    INDEX
    Explanations

    software-related terms and software versions

    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
    545
    +0.08
    0.5%
    356
    +0.07
    0.5%
    371
    +0.07
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.08
    0.24
    545
    +0.07
    0.08
    288
    +0.07
    0.22
    Negative Logits
    aucuses
    -0.44
    bing
    -0.44
    UTION
    -0.44
     tradem
    -0.43
    scill
    -0.43
    porter
    -0.43
    waters
    -0.43
    abling
    -0.42
    ills
    -0.42
    UTH
    -0.41
    POSITIVE LOGITS
    naire
    0.70
    y
    0.68
    opard
    0.68
    iasis
    0.64
    eers
    0.58
    opoulos
    0.55
    ises
    0.51
    naires
    0.50
    alis
    0.50
    pedia
    0.50
    Activations Density 9.361%

    No Known Activations