INDEX

Explanations

references to historical timelines or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

président

-0.56

setVerticalGroup

-0.53

äv

-0.46

>(&

-0.45

новништво

-0.43

 tribu

-0.43

 Sélectionnez

-0.42

 ugli

-0.42

TagHelper

-0.42

AutoresizingMask

-0.41

POSITIVE LOGITS

MigrationBuilder

0.85

 Roskov

0.72

 venons

0.64

يكب

0.63

 Wikidata

0.61

Excerpt

0.61

 kasarigan

0.60

 Spoilers

0.58

 romántico

0.58

 ✭✭

0.58

Activations Density 0.170%

references to historical timelines or events

No Comments

No Known Activations