INDEX

Explanations

mentions of important or influential individuals

oai_token-act-pair · gpt-3.5-turbo

titles or roles of prominent individuals, specifically those in leadership or professional positions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

edIn

-0.73

ilver

-0.70

 Members

-0.69

ourcing

-0.67

Bound

-0.66

english

-0.64

 Length

-0.64

etimes

-0.63

Ø§ÙĦ

-0.63

 Definitions

-0.61

POSITIVE LOGITS

 himself

0.86

's

0.78

osphere

0.77

iest

0.75

liest

0.70

 opted

0.69

stown

0.68

 laureate

0.68

 whom

0.68

owicz

0.68

Activations Density 0.294%

mentions of important or influential individuals

titles or roles of prominent individuals, specifically those in leadership or professional positions

No Comments

No Known Activations