INDEX

Explanations

phrases indicating repetitions or frequencies of actions

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

contentLoaded

-0.65

 Chwiliwch

-0.57

IBOutlet

-0.56

 مشين

-0.53

adaptiveStyles

-0.52

setVerticalGroup

-0.52

béco

-0.52

WriteTagHelper

-0.51

enumi

-0.50

IsContent

-0.50

POSITIVE LOGITS

 repeated

0.46

 thrice

0.38

>{@

0.37

repeated

0.35

Repeated

0.34

 before

0.34

 mehrm

0.33

 retries

0.33

回数

0.33

 هيا

0.32

Activations Density 0.050%

phrases indicating repetitions or frequencies of actions

No Comments

No Known Activations