INDEX

Explanations

variables or parameters related to maximum values

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "max" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

hook

-0.75

velt

-0.71

FORE

-0.69

GROUND

-0.67

 Handler

-0.66

bare

-0.65

 Patent

-0.64

 Whis

-0.64

 Athena

-0.63

RECT

-0.63

POSITIVE LOGITS

imus

1.37

imize

1.10

nesota

1.02

imil

0.97

imates

0.95

izes

0.92

imum

0.91

icum

0.87

max

0.84

icides

0.84

Activations Density 0.006%

variables or parameters related to maximum values

occurrences of the term "max" and its variations

No Comments

No Known Activations