INDEX

Explanations

phrases related to extended time periods

oai_token-act-pair · gpt-3.5-turbo

repeated references to time intervals or durations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

BI

-0.83

Guard

-0.81

LET

-0.80

clip

-0.80

ISON

-0.79

sub

-0.78

PRESS

-0.78

SU

-0.77

mie

-0.77

CHAR

-0.77

POSITIVE LOGITS

 periods

1.54

 bouts

1.29

 fluctuations

0.97

 phases

0.96

 ende

0.90

 streaks

0.90

 intervals

0.89

uits

0.87

 lapse

0.86

terness

0.84

Activations Density 0.022%

phrases related to extended time periods

repeated references to time intervals or durations

No Comments

No Known Activations