INDEX

Explanations

phrases related to rules, policies, and procedures

oai_token-act-pair · gpt-3.5-turbo

words related to rules, policies, and structures of authority

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

pload

-0.78

nen

-0.71

 Citiz

-0.71

 glim

-0.69

BUS

-0.69

ÃŃn

-0.68

plet

-0.67

ãĥĳ

-0.66

star

-0.66

 Defin

-0.65

POSITIVE LOGITS

ropy

0.73

XIV

0.65

-----

0.63

lain

0.63

 relating

0.62

":["

0.62

utra

0.61

 XIII

0.58

---

0.58

ulhu

0.58

Activations Density 0.255%

phrases related to rules, policies, and procedures

words related to rules, policies, and structures of authority

No Comments

No Known Activations