INDEX

Explanations

words related to rules, regulations, requirements, and relationships

oai_token-act-pair · gpt-3.5-turbo

words related to eligibility and restrictions within contexts

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

udeb

-0.62

 Bellev

-0.61

 Ramos

-0.60

 Aval

-0.58

 Rollins

-0.58

 Eucl

-0.57

Laf

-0.56

 Sherman

-0.54

 Kavanaugh

-0.54

 Weston

-0.54

POSITIVE LOGITS

depending

1.17

 thereof

1.06

 depending

0.88

 versa

0.87

Else

0.80

 alike

0.80

abouts

0.79

 alternatively

0.76

ancel

0.74

 guiIcon

0.74

Activations Density 0.793%

words related to rules, regulations, requirements, and relationships

words related to eligibility and restrictions within contexts

No Comments

No Known Activations