© 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. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE Local
    9. 6-RES_SCL-AJT
    11. 34691
    Prev
    Next
    INDEX
    Explanations

    descriptions of actions or events in a video game-like setting

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    211
    +0.10
    0.4%
    135
    +0.10
    0.4%
    573
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.10
    -0.10
    326
    +0.10
    0.12
    545
    +0.10
    0.04
    Negative Logits
    pmwiki
    -1.17
    EngineDebug
    -1.14
     lett
    -1.00
    ãĥ¼ãĥĨãĤ£
    -0.91
    abase
    -0.89
    ussen
    -0.87
    phalt
    -0.85
    onen
    -0.82
    orsi
    -0.82
    bytes
    -0.81
    POSITIVE LOGITS
     Okawaru
    1.14
     teleportation
    1.08
     mummy
    1.06
     Orc
    1.04
     demonic
    1.02
     Trog
    1.01
     cursed
    1.00
     Magical
    0.98
     Vault
    0.98
     Monstrous
    0.96
    Activations Density 5.516%

    No Known Activations