© 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 Local
    9. 6-RES_SCL-AJT
    11. 40490
    Prev
    Next
    INDEX
    Explanations

    phrases related to accommodation and rental properties

    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
    +0.20
    1.1%
    597
    +0.10
    0.5%
    86
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    86
    +0.20
    0.09
    554
    +0.10
    -0.20
    283
    +0.10
    -0.01
    Negative Logits
    /+
    -0.60
     Evil
    -0.53
     Correct
    -0.52
    framework
    -0.51
    Correction
    -0.50
    rid
    -0.49
    Evil
    -0.49
    Alright
    -0.49
     protocol
    -0.49
    hart
    -0.48
    POSITIVE LOGITS
     congreg
    0.73
     beware
    0.69
     experiencing
    0.66
     opting
    0.66
     clam
    0.65
    untled
    0.63
     migrate
    0.62
     desperate
    0.62
    ocumented
    0.60
     fleeing
    0.60
    Activations Density 27.198%

    No Known Activations