INDEX

Explanations

references to a subject or object in the context of actions or relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

ready

-0.49

letes

-0.45

co

-0.44

aquin

-0.44

iglio

-0.44

 Réponses

-0.43

 night

-0.43

sps

-0.43

 morning

-0.42

give

-0.41

POSITIVE LOGITS

 незавершена

0.83

tanleria

0.80

 оригіналу

0.76

twimg

0.75

InstrumentedTest

0.74

 consultato

0.72

KURZBESCHREIBUNG

0.72

TemporalType

0.72

 CascadeType

0.70

 externi

0.68

Activations Density 0.594%

references to a subject or object in the context of actions or relationships

No Comments

No Known Activations