© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 2-RES-JB
    11. 11306
    Prev
    Next
    INDEX
    Explanations

    mentions of physical or verbal attacks

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "attacked" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    inders
    -0.65
    inder
    -0.63
    Vert
    -0.62
    shown
    -0.61
    atom
    -0.61
    YC
    -0.60
    UTION
    -0.60
    soDeliveryDate
    -0.60
    cess
    -0.59
    VOL
    -0.58
    POSITIVE LOGITS
    attack
    0.96
     attacks
    0.91
    ivated
    0.84
     attack
    0.83
    oise
    0.82
     attacking
    0.82
     attackers
    0.79
    ivist
    0.76
     attacked
    0.75
     against
    0.74
    Activations Density 0.029%

    No Known Activations