INDEX

Explanations

phrases related to symbolism or importance

oai_token-act-pair · gpt-3.5-turbo

phrases that denote symbols or indicators of various concepts or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

iaries

-0.72

ade

-0.70

queue

-0.70

hari

-0.70

leys

-0.68

aceutical

-0.68

bj

-0.67

cade

-0.67

fur

-0.66

ibling

-0.63

POSITIVE LOGITS

 sorts

0.90

 discontent

0.89

 desperation

0.84

 dissatisfaction

0.83

 warmth

0.83

 masculinity

0.82

 displeasure

0.82

 patriotism

0.79

 individuality

0.79

 maturity

0.78

Activations Density 0.131%

phrases related to symbolism or importance

phrases that denote symbols or indicators of various concepts or conditions

No Comments

No Known Activations