INDEX

Explanations

words related to activities of repositioning or restarting something

oai_token-act-pair · gpt-3.5-turbo

words related to relativity or relationships

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

DragonMagazine

-0.76

HAEL

-0.73

dinand

-0.71

gerald

-0.69

 Tale

-0.67

Ĥİ

-0.65

ramid

-0.64

OLOGY

-0.64

RIC

-0.64

STEP

-0.63

POSITIVE LOGITS

abel

1.14

aunch

1.12

apsed

1.04

ibr

1.01

ocations

1.00

ocating

1.00

isted

0.99

atable

0.98

iably

0.98

aun

0.97

Activations Density 0.008%

words related to activities of repositioning or restarting something

words related to relativity or relationships

No Comments

No Known Activations