INDEX

Explanations

terms related to geographical locations or proper nouns

oai_token-act-pair · gpt-3.5-turbo

references to a specific topic involving "strat" and social media platforms, particularly Tumblr

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pedia

-0.77

ÃĽ

-0.76

rawdownloadcloneembedreportprint

-0.75

HCR

-0.70

cffffcc

-0.68

Spoiler

-0.67

utral

-0.67

phas

-0.67

 largeDownload

-0.65

hip

-0.65

POSITIVE LOGITS

OPLE

0.95

inelli

0.78

 buckle

0.76

osp

0.75

iors

0.72

rano

0.70

robe

0.70

chers

0.67

vier

0.66

agne

0.66

Activations Density 0.112%

terms related to geographical locations or proper nouns

references to a specific topic involving "strat" and social media platforms, particularly Tumblr

No Comments

No Known Activations