© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 8-RES-JB
    6. 4916
    Prev
    Next
    INDEX
    Explanations

    phrases related to removal or elimination

    oai_token-act-pair · gpt-3.5-turbo

    recurrent references to the concept of removal or elimination

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    step
    -0.75
    ÃĥÃĤ
    -0.74
    CLA
    -0.73
     Finally
    -0.71
    yssey
    -0.70
    WM
    -0.70
    CLASSIFIED
    -0.69
    SHIP
    -0.69
    é¾įåĸļ士
    -0.69
    iculture
    -0.69
    POSITIVE LOGITS
     pesky
    1.16
     possibility
    1.05
     altogether
    1.04
     offending
    1.03
     unnecessary
    1.00
     veil
    0.99
     superflu
    0.99
     redundant
    0.98
     entirety
    0.97
     obsolete
    0.92
    Activations Density 0.243%

    No Known Activations