© 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. 3841
    Prev
    Next
    INDEX
    Explanations

    situational conditions or circumstances

    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
    288
    +0.16
    0.5%
    715
    +0.16
    0.5%
    457
    +0.15
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.16
    0.18
    540
    +0.16
    0.16
    324
    +0.15
    0.10
    Negative Logits
    apses
    -1.28
    assies
    -1.20
    azines
    -1.16
    andise
    -1.08
    vironments
    -1.07
    apons
    -1.05
    events
    -1.04
    sels
    -0.98
    endas
    -0.97
    uctions
    -0.96
    POSITIVE LOGITS
     unto
    1.11
    brainer
    1.09
     magnet
    1.06
    less
    1.03
     exaggeration
    1.00
     reminder
    1.00
     impossibility
    0.98
     worthy
    0.97
     approximation
    0.95
     understatement
    0.93
    Activations Density 3.992%

    No Known Activations