INDEX

Explanations

words related to celestial bodies, particularly "Stars."

oai_token-act-pair · gpt-3.5-turbo

references to the word "Stars" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

apons

-0.71

ibaba

-0.71

vice

-0.68

ufact

-0.67

armac

-0.67

 Parkinson

-0.67

mediate

-0.66

senal

-0.65

psons

-0.65

oad

-0.64

POSITIVE LOGITS

 Stars

1.34

Stars

1.23

cream

1.17

stars

1.03

hips

0.90

 Devils

0.84

bucks

0.82

 stars

0.81

 Catalog

0.80

ystem

0.79

Activations Density 0.012%

words related to celestial bodies, particularly "Stars."

references to the word "Stars" in various contexts

No Comments

No Known Activations