INDEX

Explanations

terms related to data structuring and computational processes

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

specific words or punctuation

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

ANTLR

-0.39

sbericht

-0.35

 AssemblyTitle

-0.32

OK

-0.31

ค้า

-0.30

ياء

-0.30

 okay

-0.29

 Syrie

-0.29

 sideways

-0.29

 opérateur

-0.29

POSITIVE LOGITS

PerformLayout

0.82

enderror

0.65

AsUp

0.62

RenderAtEndOf

0.62

aarrggbb

0.60

GraphicsUnit

0.60

ListItemIcon

0.57

 NSCoder

0.56

intios

0.56

NewUrlParser

0.56

Activations Density 0.139%

terms related to data structuring and computational processes

specific words or punctuation

No Comments

No Known Activations