INDEX

Explanations

mentions of characters or people in a text

oai_token-act-pair · gpt-3.5-turbo

references to characters in narratives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

VERTIS

-0.72

rup

-0.72

×ª

-0.72

yg

-0.70

ntil

-0.67

LOCK

-0.66

Effective

-0.66

condition

-0.64

galitarian

-0.63

 Tant

-0.63

POSITIVE LOGITS

acters

1.52

istically

1.17

istics

1.07

 arcs

0.86

izations

0.86

 characters

0.83

 portraits

0.82

 assassinate

0.82

isations

0.78

 Characters

0.78

Activations Density 0.046%

mentions of characters or people in a text

references to characters in narratives

No Comments

No Known Activations