Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 8788
    Prev
    Next
    INDEX
    Explanations

    phrases related to being physically away or off from a specific location or task

    oai_token-act-pair · gpt-3.5-turbo

    expressions indicating distance or separation from a place or situation

    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
    IRE
    -0.69
     Spur
    -0.66
     Dame
    -0.64
     purported
    -0.59
     Shoot
    -0.58
     CTR
    -0.58
    Sequ
    -0.58
     superflu
    -0.56
     mosaic
    -0.56
    Else
    -0.55
    POSITIVE LOGITS
    enced
    0.91
    fed
    0.86
    itable
    0.86
    pport
    0.85
    loaded
    0.84
    ranged
    0.79
    raged
    0.78
    active
    0.78
    itored
    0.77
    lined
    0.76
    Activations Density 0.188%

    No Known Activations