INDEX

Explanations

references to organizations, political entities, and collaborations related to communication and cultural studies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 itſelf

-1.16

ſelves

-1.09

 themſelves

-1.08

 myſelf

-1.08

 Monfieur

-1.08

^(@)

-1.05

ſelf

-1.04

 Majefty

-1.04

 Reſ

-1.01

 Anſ

-1.01

POSITIVE LOGITS

plus

0.64

0.59

+.

0.58

label

0.57

+,

0.55

 Concept

0.53

Concept

0.51

 Plus

0.50

enderror

0.50

Activations Density 0.231%

references to organizations, political entities, and collaborations related to communication and cultural studies

No Comments

No Known Activations