INDEX

Explanations

phrases related to starting or beginning something

oai_token-act-pair · gpt-3.5-turbo

phrases related to the beginning or initiation of actions or processes

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

lain

-0.75

elong

-0.65

 integrity

-0.64

abled

-0.63

ointed

-0.62

luence

-0.59

Crime

-0.58

abytes

-0.58

 dearly

-0.57

safety

-0.57

POSITIVE LOGITS

Uni

0.73

CG

0.69

imester

0.68

itially

0.67

HK

0.66

ahime

0.64

NG

0.63

 Paragu

0.62

©¶æ¥µ

0.62

 Micha

0.61

Activations Density 0.038%

phrases related to starting or beginning something

phrases related to the beginning or initiation of actions or processes

No Comments

No Known Activations