INDEX

Explanations

sentences starting with specific characters (like dots) followed by various characters that don't adhere to a clear pattern

oai_token-act-pair · gpt-3.5-turbo

references to specific characters or roles in narrative contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 mosqu

-0.76

 democrat

-0.75

pport

-0.73

 urgent

-0.73

 congest

-0.72

arching

-0.71

 decentral

-0.70

hene

-0.70

 herb

-0.67

 consolidate

-0.67

POSITIVE LOGITS

Trivia

1.45

Appearances

1.44

Gallery

1.32

Contents

1.26

Quotes

1.18

Characters

1.17

Episode

1.15

episode

1.15

Season

1.11

 Played

1.08

Activations Density 0.347%

sentences starting with specific characters (like dots) followed by various characters that don't adhere to a clear pattern

references to specific characters or roles in narrative contexts

No Comments

No Known Activations