INDEX

Explanations

phrases indicating complex situations with an underlying sense of irony or critique

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 alent

-0.50

FORMANCE

-0.49

 Schulte

-0.49

чины

-0.49

μαν

-0.48

herjee

-0.47

OON

-0.46

 Newell

-0.45

 holding

-0.45

település

-0.45

POSITIVE LOGITS

RegressionTest

0.78

 Gump

0.69

WriteTagHelper

0.68

 дописавши

0.67

 MaterialApp

0.65

ValueGeneration

0.64

errHandler

0.62

 ModelExpression

0.60

Sucesor

0.59

Portale

0.59

Activations Density 0.225%

phrases indicating complex situations with an underlying sense of irony or critique

No Comments

No Known Activations