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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 9-RES-JB
    6. 7693
    Prev
    Next
    INDEX
    Explanations

    materials or substances that something is made of

    oai_token-act-pair · gpt-3.5-turbo

    phrases that describe materials used in making objects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    osponsors
    -0.85
    Sexual
    -0.73
    Interstitial
    -0.72
    IRC
    -0.71
     challeng
    -0.71
    taboola
    -0.69
    isSpecialOrderable
    -0.69
    Apps
    -0.69
    license
    -0.68
    isode
    -0.68
    POSITIVE LOGITS
     recycled
    1.18
     materials
    1.14
     scratch
    1.07
     wood
    0.99
     pure
    0.97
     scraps
    0.96
     steel
    0.92
     aluminum
    0.92
     atoms
    0.91
     clay
    0.91
    Activations Density 0.110%

    No Known Activations