INDEX

Explanations

phrases related to timing or sequential events

oai_token-act-pair · gpt-3.5-turbo

temporal phrases that indicate sequence or timing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

icter

-0.76

âĶĢ

-0.67

lesh

-0.64

hemer

-0.64

subject

-0.62

killed

-0.62

imilation

-0.62

locks

-0.61

ritten

-0.61

Registered

-0.61

POSITIVE LOGITS

ĪĴ

0.76

 reinforcements

0.64

 pursuit

0.64

 laughter

0.63

 toile

0.62

odo

0.61

enium

0.60

 sheer

0.60

 regard

0.59

 corridors

0.59

Activations Density 0.551%

phrases related to timing or sequential events

temporal phrases that indicate sequence or timing

No Comments

No Known Activations