INDEX

Explanations

indicators of an indeterminate or unspecified format or structure in code or text

oai_token-act-pair · gpt-4o-mini Triggered by @bot

subscription query

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

AccessorTable

-0.81



-0.73

 propOrder

-0.73

awtextra

-0.69

RegressionTest

-0.68

fromnode

-0.68

UseVisualStyle

-0.67

ResumeLayout

-0.66

 ExecuteAsync

-0.65

 <=",

-0.65

POSITIVE LOGITS

}{*}{}

0.53

}{*}{

0.37

ammans

0.36

 âmes

0.35

the

0.34

 esprits

0.34

ReusableCell

0.34

0.33

 bruja

0.33

Còn

0.32

Activations Density 0.002%

indicators of an indeterminate or unspecified format or structure in code or text

subscription query

No Comments

No Known Activations