© 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. 415
    Prev
    Next
    INDEX
    Explanations

    phrases related to locations or organizations with the word "Rec" in them

    oai_token-act-pair · gpt-3.5-turbo

    references to "Recreation" or "Recreational" activities

    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
    heid
    -0.85
    wit
    -0.81
    agra
    -0.80
     fuck
    -0.74
     tits
    -0.74
    rael
    -0.72
    fuck
    -0.71
     bitch
    -0.71
     fucking
    -0.67
    gay
    -0.66
    POSITIVE LOGITS
     Rec
    3.61
    Rec
    2.12
     rec
    1.78
    rec
    1.67
     REC
    1.64
     Recreation
    1.37
     Recon
    1.25
    REC
    1.24
     Recre
    1.20
     recy
    1.19
    Activations Density 0.024%

    No Known Activations