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. 11-RES-JB
    6. 15631
    Prev
    Next
    INDEX
    Explanations

    verbs indicating purpose or intention

    oai_token-act-pair · gpt-3.5-turbo

    expressions of intent or desire to change or act upon social structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    mails
    -0.76
    shots
    -0.75
     Caller
    -0.74
     Parking
    -0.73
    Got
    -0.72
     Garr
    -0.68
    glers
    -0.67
     Bucket
    -0.67
    luck
    -0.66
    listed
    -0.65
    POSITIVE LOGITS
     abolish
    1.50
     liberate
    1.38
     emanc
    1.35
     embody
    1.31
     reconcile
    1.30
     impose
    1.29
     overthrow
    1.28
     eradicate
    1.28
     redef
    1.28
     transcend
    1.27
    Activations Density 0.205%

    No Known Activations