© 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. 10-RES-JB
    11. 12914
    Prev
    Next
    INDEX
    Explanations

    mentions of victims or victorious events

    oai_token-act-pair · gpt-3.5-turbo

    terms related to victims and victimization

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Shank
    -0.70
    backer
    -0.68
     waivers
    -0.68
     grounds
    -0.63
    WAYS
    -0.63
    Factor
    -0.60
    Kin
    -0.59
     Hath
    -0.59
     reservation
    -0.59
     marrow
    -0.58
    POSITIVE LOGITS
    orian
    1.73
    orious
    1.63
    oire
    1.58
    ory
    1.57
    orians
    1.54
    oria
    1.53
    orical
    1.48
    orius
    1.39
    ories
    1.38
    oric
    1.37
    Activations Density 0.035%

    No Known Activations