INDEX

Explanations

phrases or words that have a high level of association or similarity

oai_token-act-pair · gpt-3.5-turbo

terms related to equivalence and interconnectedness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

annis

-0.71

GER

-0.68

GRE

-0.67

adra

-0.67

MRI

-0.65

vae

-0.64

UGE

-0.62

Doctors

-0.62

ARM

-0.60

uld

-0.60

POSITIVE LOGITS

 synonymous

1.30

icative

0.85

entimes

0.84

onymous

0.83

inational

0.80

uously

0.78

omorph

0.77

onym

0.73

omorphic

0.73

 Nadu

0.71

Activations Density 0.013%

phrases or words that have a high level of association or similarity

terms related to equivalence and interconnectedness

No Comments

No Known Activations