Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 4170
    Prev
    Next
    INDEX
    Explanations

    sports terms and names

    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
    447
    +0.28
    1.6%
    146
    +0.11
    0.6%
    296
    +0.08
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    146
    +0.28
    0.16
    442
    +0.11
    0.13
    179
    +0.08
    0.11
    Negative Logits
     EVE
    -0.58
    taboola
    -0.55
     Syri
    -0.54
     Mulcair
    -0.51
     PKK
    -0.51
     Sven
    -0.49
     ç¥ŀ
    -0.48
     Pyrrha
    -0.48
    adow
    -0.48
     ;;
    -0.48
    POSITIVE LOGITS
     Headquarters
    0.68
     Commissioner
    0.65
     headquarters
    0.64
     offices
    0.63
     regulations
    0.60
     Chairman
    0.59
     commissioners
    0.58
     Executive
    0.57
     bureaucracy
    0.57
    's
    0.57
    Activations Density 14.278%

    No Known Activations