INDEX

Explanations

proper nouns and organizations related to politics, academia, and possibly specific regions

oai_token-act-pair · gpt-3.5-turbo

references to specific names, particularly related to notable individuals or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

holders

-1.04

holder

-0.82

mble

-0.72

erness

-0.71

itures

-0.70

rights

-0.69

ages

-0.68

hetical

-0.67

comings

-0.66

igators

-0.66

POSITIVE LOGITS

upiter

1.01

unction

1.00

oint

0.95

ournals

0.93

utsu

0.91

igsaw

0.88

ealous

0.88

avascript

0.86

ernaut

0.86

ihad

0.84

Activations Density 0.094%

proper nouns and organizations related to politics, academia, and possibly specific regions

references to specific names, particularly related to notable individuals or entities

No Comments

No Known Activations