INDEX

Explanations

items or actions related to increasing something

oai_token-act-pair · gpt-3.5-turbo

instances of the word "the" and contexts indicating an increase in quantity or risk

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

books

-0.76

mong

-0.74

craft

-0.67

mates

-0.66

mate

-0.65

SPONSORED

-0.65

picked

-0.64

berg

-0.64

 Xuan

-0.64

dig

-0.64

POSITIVE LOGITS

 burden

0.95

 likelihood

0.93

 severity

0.92

 intensity

0.91

 competitiveness

0.91

 amount

0.89

 incidence

0.88

 overall

0.87

 effectiveness

0.86

 possibility

0.86

Activations Density 0.226%

items or actions related to increasing something

instances of the word "the" and contexts indicating an increase in quantity or risk

No Comments

No Known Activations