INDEX

Explanations

significant numbers related to events or quantities

oai_token-act-pair · gpt-3.5-turbo

quantitative data related to incidents and events

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

amaru

-0.73

FontSize

-0.68

uran

-0.67

 slogan

-0.65

atown

-0.65

ãĥ¼ãĥ«

-0.64

opia

-0.63

imaru

-0.62

shape

-0.62

cription

-0.62

POSITIVE LOGITS

 consecutive

1.02

 instances

1.01

0.97

 percent

0.91

teen

0.89

 cases

0.88

 separate

0.81

 occasions

0.80

 deaths

0.78

 occurrences

0.77

Activations Density 0.270%

significant numbers related to events or quantities

quantitative data related to incidents and events

No Comments

No Known Activations