© 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. 4158
    Prev
    Next
    INDEX
    Explanations

    terms related to the genre of reggae music

    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
    644
    +0.14
    0.7%
    447
    +0.12
    0.6%
    503
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.14
    0.14
    507
    +0.12
    0.08
    251
    +0.12
    0.05
    Negative Logits
    hower
    -0.81
    lihood
    -0.76
    WARE
    -0.73
     Hole
    -0.70
     Trap
    -0.69
    BOOK
    -0.68
     Fey
    -0.68
    ÙIJ
    -0.68
    shapeshifter
    -0.65
     Traps
    -0.65
    POSITIVE LOGITS
    ressive
    1.10
    roup
    1.08
    inal
    1.08
    rowth
    1.06
    aining
    1.04
    ents
    1.02
    nant
    1.01
    ency
    1.00
    rett
    0.98
    iments
    0.97
    Activations Density 4.148%

    No Known Activations