© 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. 4-RES-JB
    11. 10499
    Prev
    Next
    INDEX
    Explanations

    information related to game abilities or effects

    oai_token-act-pair · gpt-3.5-turbo

    metrics and effects related to movement and damage in a gaming context

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

    No Comments

    Negative Logits
    Reviewer
    -0.86
    çĶ°
    -0.85
     è£ıè
    -0.77
    hett
    -0.76
     Homo
    -0.75
     handwriting
    -0.75
     guiActiveUn
    -0.73
     horizont
    -0.73
     Sach
    -0.73
    sburg
    -0.73
    POSITIVE LOGITS
     cooldown
    1.21
     Cooldown
    1.09
     dmg
    1.06
     projectiles
    1.05
    Cooldown
    1.05
    Spell
    1.01
    Damage
    1.00
     Damage
    1.00
    Weapon
    0.99
     melee
    0.94
    Activations Density 0.193%

    No Known Activations