© 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. 6146
    Prev
    Next
    INDEX
    Explanations

    values or numbers indicated with a plus or minus sign

    oai_token-act-pair · gpt-3.5-turbo

    negative values or terms indicating negativity

    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
     Breed
    -0.62
     ambitious
    -0.61
    urg
    -0.61
     Brist
    -0.58
     Frie
    -0.56
     Bristol
    -0.56
     investigations
    -0.56
    ingham
    -0.56
    mong
    -0.56
     investigation
    -0.55
    POSITIVE LOGITS
     minus
    3.93
    minus
    3.52
     plus
    1.67
     +/-
    1.24
     subtract
    1.15
    plus
    1.14
     PLUS
    1.07
    excluding
    1.00
     sans
    1.00
    renheit
    0.99
    Activations Density 0.010%

    No Known Activations