INDEX

Explanations

names in text

oai_token-act-pair · gpt-3.5-turbo

links to social media content or mentions of user engagement

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

anamo

-0.80

sic

-0.76

ursday

-0.73

ryu

-0.73

odon

-0.72

onte

-0.70

etheless

-0.70

inki

-0.70

obin

-0.69

ichick

-0.69

POSITIVE LOGITS

Quantity

0.80

Description

0.76

Overview

0.76

bread

0.75

Contact

0.74

GROUND

0.74

Development

0.73

Disclaimer

0.71

 ILCS

0.70

Rate

0.70

Activations Density 0.061%

names in text

links to social media content or mentions of user engagement

No Comments

No Known Activations