INDEX

Explanations

phrases related to breaking news and newsletters

oai_token-act-pair · gpt-3.5-turbo

references to breaking news

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

oka

-0.69

ngth

-0.66

 racially

-0.65

 fart

-0.63

 phys

-0.62

 discriminated

-0.62

ONSORED

-0.62

 chilly

-0.62

 dunno

-0.61

 misunder

-0.61

POSITIVE LOGITS

Desk

0.84

 Emails

0.79

circ

0.74

Trend

0.72

Dispatch

0.70

alion

0.69

cycle

0.68

aucus

0.67

fusc

0.67

line

0.66

Activations Density 0.017%

phrases related to breaking news and newsletters

references to breaking news

No Comments

No Known Activations