INDEX

Explanations

the word "pres" followed by a number, possibly indicating precision or prescription

oai_token-act-pair · gpt-3.5-turbo

references to prescriptive or authoritative guidance

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

TAIN

-0.83

cles

-0.77

ress

-0.72

ISM

-0.69

nered

-0.67

lett

-0.66

itarian

-0.65

Bear

-0.65

ruary

-0.64

beans

-0.63

POSITIVE LOGITS

wow

0.82

erver

0.79

 pres

0.79

ervative

0.77

acc

0.74

olk

0.74

ider

0.70

ommel

0.70

inos

0.69

rontal

0.69

Activations Density 0.029%

the word "pres" followed by a number, possibly indicating precision or prescription

references to prescriptive or authoritative guidance

No Comments

No Known Activations