INDEX

Explanations

terms related to news stories or headlines

oai_token-act-pair · gpt-3.5-turbo

references to media sources and types of content

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

atri

-0.68

thro

-0.64

oint

-0.64

 strengths

-0.62

Oracle

-0.62

 order

-0.59

broken

-0.59

bows

-0.58

 bleeding

-0.58

lag

-0.58

POSITIVE LOGITS

 IMAGES

1.88

 WATCHED

0.99

 PHOTO

0.94

URE

0.89

 VIDEOS

0.82

 POLIT

0.77

 âĢº

0.76

TY

0.76

URES

0.76

 Images

0.75

Activations Density 0.016%

terms related to news stories or headlines

references to media sources and types of content

No Comments

No Known Activations