© 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. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 0-RES-JB
    11. 20541
    Prev
    Next
    INDEX
    Explanations

    phrases related to repetitions or multiples

    oai_token-act-pair · gpt-3.5-turbo

    references to "triple" and its relationship to quantities or comparisons

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    anners
    -0.79
     POLIT
    -0.69
    Auth
    -0.67
    ila
    -0.66
     Painter
    -0.66
    beit
    -0.64
    lang
    -0.64
    eller
    -0.63
     Fein
    -0.63
    owsky
    -0.63
    POSITIVE LOGITS
     triple
    3.55
     quadru
    2.33
     Triple
    2.26
     double
    1.66
     tripled
    1.51
    tri
    1.49
    Tri
    1.46
     tri
    1.31
     Double
    1.29
    Double
    1.21
    Activations Density 0.016%

    No Known Activations