© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 2473
    Prev
    Next
    INDEX
    Explanations

    phrases indicating direction or progress towards a particular goal or endpoint

    oai_token-act-pair · gpt-3.5-turbo

    various contexts where the word "towards" is used

    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
     Edison
    -0.63
    essions
    -0.63
     abbrevi
    -0.61
     Plain
    -0.58
     Kidd
    -0.57
     shutdown
    -0.57
     Recall
    -0.57
     FEMA
    -0.55
    undown
    -0.55
    und
    -0.54
    POSITIVE LOGITS
     towards
    3.45
     toward
    2.91
     Towards
    2.45
     downwards
    1.70
     upwards
    1.28
     onwards
    1.28
     Tow
    1.14
    oward
    1.14
     whilst
    1.13
    wards
    1.12
    Activations Density 0.025%

    No Known Activations