INDEX

Explanations

numerical values identifying the maximum or limit of something

oai_token-act-pair · gpt-3.5-turbo

the term "max" related to maximum values or limits

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

FORE

-0.85

perse

-0.69

 Nare

-0.69

 Journalism

-0.67

 Parkinson

-0.67

 Forth

-0.66

GROUND

-0.65

Interstitial

-0.64

hook

-0.63

 Kong

-0.62

POSITIVE LOGITS

imus

1.14

nesota

1.07

imize

1.01

imates

0.95

izes

0.88

icum

0.88

ima

0.88

ims

0.87

max

0.84

egal

0.82

Activations Density 0.011%

numerical values identifying the maximum or limit of something

the term "max" related to maximum values or limits

No Comments

No Known Activations