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. 21177
    Prev
    Next
    INDEX
    Explanations

    words related to counter-intelligence and counter-terrorism

    oai_token-act-pair · gpt-3.5-turbo

    terms related to counterintelligence and counterterrorism

    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
     Indigo
    -0.72
     monog
    -0.69
     Norn
    -0.69
     Ago
    -0.65
     Phi
    -0.65
    entric
    -0.64
     squid
    -0.64
     laure
    -0.63
     PVC
    -0.62
     abstinence
    -0.61
    POSITIVE LOGITS
    terrorism
    1.19
    measures
    1.17
    intuitive
    1.14
    strike
    1.09
    attack
    1.08
    intelligence
    1.08
    matter
    1.05
    argument
    1.01
    campaign
    1.01
    productive
    1.00
    Activations Density 0.050%

    No Known Activations