INDEX

Explanations

phrases and structures related to definitions and descriptions within technical contexts

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

description of identifiers

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

GOTREF

-0.43

shtml

-0.34

processable

-0.30

 policiers

-0.29

mailto

-0.29

://$

-0.29

']>;

-0.29

 techniczne

-0.29

]=>

-0.28

站

-0.28

POSITIVE LOGITS

ſammen

0.63

 lenker

0.63

rungsseite

0.60

UnusedPrivate

0.59

 '\\;'

0.56

 betweenstory

0.54

itinéraire

0.52

tvguidetime

0.50

MemoryWarning

0.49

 queſto

0.49

Activations Density 0.039%

phrases and structures related to definitions and descriptions within technical contexts

description of identifiers

No Comments

No Known Activations