INDEX

Explanations

the word "as" used in various contexts, typically indicating comparison or function, along with instances of "at" related to locations or positions

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

at (coordinate)

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

ReusableCell

-0.92

 queſta

-0.90

 OMITBAD

-0.85

Tembelea

-0.84

AddTagHelper

-0.83

transQ

-0.81

expandindo

-0.79

tvguidetime

-0.77

UserScript

-0.77

tagHelperRunner

-0.76

POSITIVE LOGITS

↵↵

0.54

for

0.51

 from

0.48

by

0.47

↵

0.47

↵↵↵

0.46

In

0.45

By

0.45

For

0.45

if

0.45

Activations Density 0.113%

the word "as" used in various contexts, typically indicating comparison or function, along with instances of "at" related to locations or positions

at (coordinate)

No Comments

No Known Activations