© 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. 3885
    Prev
    Next
    INDEX
    Explanations

    titles of songs and albums that are being remixed

    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
    712
    +0.13
    0.5%
    288
    +0.11
    0.4%
    131
    +0.11
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.13
    0.14
    346
    +0.11
    0.07
    161
    +0.11
    0.05
    Negative Logits
    iencies
    -0.92
     Liang
    -0.84
     Scal
    -0.80
    osate
    -0.79
    anmar
    -0.76
    htaking
    -0.76
     Govern
    -0.76
     Gand
    -0.76
     retri
    -0.75
    rir
    -0.75
    POSITIVE LOGITS
     lyrics
    1.70
     lyric
    1.46
     albums
    1.43
     album
    1.40
     drummer
    1.38
     vocals
    1.37
     songs
    1.36
     music
    1.35
     Songs
    1.32
    music
    1.32
    Activations Density 4.016%

    No Known Activations