INDEX

Explanations

publication dates

oai_token-act-pair · gpt-3.5-turbo

instances of the word "published" along with associated timestamps

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

utic

-0.82

porary

-0.81

adra

-0.74

vette

-0.71

otropic

-0.71

chenko

-0.71

ixel

-0.70

oise

-0.69

antics

-0.69

adr

-0.68

POSITIVE LOGITS

 aloud

0.79

 Date

0.75

 Prediction

0.71

Apr

0.71

Published

0.69

 Stories

0.67

 Decision

0.66

NESS

0.65

 mang

0.64

 behavi

0.64

Activations Density 0.018%

publication dates

instances of the word "published" along with associated timestamps

No Comments

No Known Activations