INDEX

Explanations

mathematical expressions and programming code snippets

oai_token-act-pair · gpt-3.5-turbo

symbols or notations that indicate mathematical or logical operations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

anium

-0.79

essee

-0.79

 Grimes

-0.73

 Lauder

-0.72

owicz

-0.71

itzer

-0.71

xton

-0.71

NER

-0.67

vey

-0.65

iger

-0.62

POSITIVE LOGITS

/-

1.19

-+

1.08

/+

0.90

-+-+-+-+

0.83

aign

0.82

--+

0.80

dash

0.79

raid

0.78

ssh

0.76

Pg

0.75

Activations Density 0.013%

mathematical expressions and programming code snippets

symbols or notations that indicate mathematical or logical operations

No Comments

No Known Activations