© 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. 4-RES-JB
    11. 6106
    Prev
    Next
    INDEX
    Explanations

    names of individuals or organizations

    oai_token-act-pair · gpt-3.5-turbo

    historical events and key dates

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     emoji
    -0.82
     Snapchat
    -0.80
     (@
    -0.76
     POLITICO
    -0.72
    ðŁ
    -0.70
    ï¸ı
    -0.70
     tweeting
    -0.69
     ðŁij
    -0.68
     ðŁ
    -0.68
     IoT
    -0.68
    POSITIVE LOGITS
     Yugoslav
    0.95
     postwar
    0.95
    oslov
    0.91
     Yugoslavia
    0.90
     inval
    0.82
     Moroc
    0.82
     Soviet
    0.80
     colonial
    0.79
     UNESCO
    0.75
     historians
    0.75
    Activations Density 2.408%

    No Known Activations