© 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 E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 68
    Prev
    Next
    INDEX
    Explanations

    photos or images and possibly associated details like captions

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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
    430
    +0.16
    0.5%
    194
    +0.13
    0.4%
    268
    +0.13
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    387
    +0.16
    0.03
    710
    +0.13
    0.02
    739
    +0.13
    0.02
    Negative Logits
    adden
    -1.35
    cffff
    -1.12
    imore
    -1.04
    iazep
    -1.04
    oplan
    -1.02
    ername
    -1.02
    ylum
    -1.01
    uckle
    -1.00
    hetamine
    -0.99
    hon
    -0.98
    POSITIVE LOGITS
     Gallery
    1.06
     Artifact
    1.00
    grounds
    0.98
     gallery
    0.93
    ifacts
    0.92
     Nerd
    0.91
     shutter
    0.90
     galleries
    0.90
     beads
    0.89
    pedia
    0.89
    Activations Density 0.062%

    No Known Activations