© 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. 3-RES-JB
    11. 10713
    Prev
    Next
    INDEX
    Explanations

    phrases related to sponsorship or financial support

    oai_token-act-pair · gpt-3.5-turbo

    terms related to financial support and backing

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Hallow
    -0.71
     coroner
    -0.70
     Milky
    -0.69
     skelet
    -0.68
     Ches
    -0.65
    ogram
    -0.62
     Hamp
    -0.62
     Iw
    -0.62
    ©¶æ¥µ
    -0.61
     Ingredients
    -0.60
    POSITIVE LOGITS
    ablishment
    0.92
     persecuted
    0.86
    Supporters
    0.78
    isance
    0.76
    itism
    0.75
     persecut
    0.74
     justifies
    0.74
     deportation
    0.73
     appease
    0.72
     sanctions
    0.71
    Activations Density 0.301%

    No Known Activations