INDEX

Explanations

words related to personal opinions or declarations

oai_token-act-pair · gpt-3.5-turbo

phrases related to statements of denial or negation

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

alore

-0.69

ammy

-0.66

iage

-0.65

 fortun

-0.64

 pressures

-0.64

 Gutenberg

-0.63

 scattering

-0.62

anmar

-0.62

 presumed

-0.61

pse

-0.60

POSITIVE LOGITS

ï¸ı

0.86

İ

0.82

VICE

0.80

hall

0.78

ï¸

0.76

hood

0.74

worthiness

0.73

ÙĦ

0.72

Balt

0.72

sure

0.72

Activations Density 0.293%

words related to personal opinions or declarations

phrases related to statements of denial or negation

No Comments

No Known Activations