© 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. 8-RES-JB
    11. 3227
    Prev
    Next
    INDEX
    Explanations

    verbs related to actions or abilities

    oai_token-act-pair · gpt-3.5-turbo

    indications of actions or states relating to performance or capabilities

    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
    ð
    -0.75
    bush
    -0.66
     Verse
    -0.66
    mpire
    -0.65
    thal
    -0.65
    whe
    -0.63
    conn
    -0.63
    ire
    -0.63
    pal
    -0.63
    ry
    -0.61
    POSITIVE LOGITS
     moreover
    1.24
     furthermore
    1.22
     also
    1.18
     therefore
    1.06
     however
    0.99
     meanwhile
    0.91
     additionally
    0.90
     certainly
    0.85
     thus
    0.79
    also
    0.76
    Activations Density 0.531%

    No Known Activations