INDEX

Explanations

words related to functions that modify or change values

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

mutating func, mutated by, molten metal

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

 betweenstory

-0.67

transQ

-0.63

 الحره

-0.60

ckså

-0.57

findpost

-0.56

Tembelea

-0.56

verwijspagina

-0.55

windowFixed

-0.54

 itſelf

-0.54

 tričko

-0.53

POSITIVE LOGITS

mut

0.85

Mut

0.72

mut

0.71

Mut

0.56

MUT

0.53

 mergeFrom

0.51

 mutate

0.49

 mutant

0.47

 mutated

0.47

posedge

0.44

Activations Density 0.003%

words related to functions that modify or change values

mutating func, mutated by, molten metal

No Comments

No Known Activations