© 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. 3669
    Prev
    Next
    INDEX
    Explanations

    references to family relationships and ownership

    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
    223
    +0.16
    0.5%
    50
    +0.15
    0.5%
    288
    +0.14
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.16
    0.09
    711
    +0.15
    0.05
    485
    +0.14
    0.06
    Negative Logits
    orsi
    -1.26
    atility
    -1.26
    anwhile
    -1.25
    cale
    -1.20
    cape
    -1.11
    ourning
    -1.06
    schild
    -1.02
    haw
    -1.02
    pite
    -1.01
    paio
    -1.01
    POSITIVE LOGITS
    hesis
    1.55
    heses
    1.46
    hetical
    1.44
    hood
    1.12
    hetically
    1.07
    parts
    1.02
    ship
    0.98
    mast
    0.97
    States
    0.95
     countries
    0.93
    Activations Density 1.019%

    No Known Activations