INDEX

Explanations

the term "concept" and its variations in different contexts

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

ásban

-0.64

WriteTo

-0.61

abon

-0.58

 hår

-0.57

ásra

-0.55

ri

-0.54

FromArgb

-0.54

 resistenza

-0.53

 cser

-0.53

on

-0.52

POSITIVE LOGITS

 concepts

2.40

 concept

2.40

 Concept

2.13

Concept

2.10

 CONCEPT

2.10

concept

2.09

 Concepts

2.04

concepts

1.95

Concepts

1.93

 CONCEPTS

1.75

Activations Density 0.069%

the term "concept" and its variations in different contexts

No Comments

No Known Activations