INDEX

Explanations

phrases related to location or direction

oai_token-act-pair · gpt-3.5-turbo

inquiries about location or presence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

yi

-0.83

ATURE

-0.73

ÂłÂłÂłÂłÂłÂłÂłÂł

-0.68

vous

-0.68

astics

-0.65

asts

-0.65

wave

-0.64

rolet

-0.64

ME

-0.62

SPONSORED

-0.62

POSITIVE LOGITS

abouts

1.38

upon

1.11

fore

1.01

soever

0.80

ver

0.72

 else

0.72

with

0.69

Dat

0.66

ipl

0.66

 exactly

0.66

Activations Density 0.054%

phrases related to location or direction

inquiries about location or presence

No Comments

No Known Activations