© 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
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE Local
    5. 6-RES_SCL-AJT
    6. 34845
    Prev
    Next
    INDEX
    Explanations

    mentions of aim or aiming

    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
    151
    +0.10
    0.5%
    557
    +0.09
    0.5%
    643
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    549
    +0.10
    0.06
    439
    +0.09
    0.04
    239
    +0.09
    0.06
    Negative Logits
    natureconservancy
    -0.73
    kees
    -0.58
    thia
    -0.57
    meal
    -0.57
    SPONSORED
    -0.56
    kered
    -0.55
    NetMessage
    -0.54
    $$$$
    -0.53
    hander
    -0.51
     liquids
    -0.50
    POSITIVE LOGITS
    onica
    0.76
     Saban
    0.73
    stadt
    0.73
    imum
    0.72
    ulus
    0.70
    eer
    0.69
    arro
    0.68
    ento
    0.68
    imus
    0.66
    atar
    0.66
    Activations Density 6.257%

    No Known Activations