INDEX

Explanations

phrases indicating a negative perception or judgment of a situation

oai_token-act-pair · gpt-3.5-turbo

the phrase "all too" in various contexts of overuse or excess

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

agara

-0.78

ahime

-0.77

ersion

-0.71

utenberg

-0.71

cius

-0.70

agra

-0.68

cheat

-0.65

ushi

-0.63

EStream

-0.62

agna

-0.62

POSITIVE LOGITS

ÏĤ

0.64

Reloaded

0.62

~~~~~~~~

0.62

SAM

0.59

lihood

0.58

nuts

0.57

 omin

0.55

Todd

0.54

Round

0.54

ira

0.53

Activations Density 0.045%

phrases indicating a negative perception or judgment of a situation

the phrase "all too" in various contexts of overuse or excess

No Comments

No Known Activations