INDEX

Explanations

phrases related to suggesting or evaluating actions

oai_token-act-pair · gpt-3.5-turbo

phrases that express good ideas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

anches

-0.74

cow

-0.69

wood

-0.69

 trophies

-0.68

abbling

-0.68

roads

-0.68

cki

-0.68

heast

-0.67

DF

-0.65

builders

-0.65

POSITIVE LOGITS

ually

1.01

 proposition

0.70

 moot

0.69

urally

0.64

meal

0.64

ural

0.64

ACY

0.64

HDL

0.63

DonaldTrump

0.63

Mob

0.61

Activations Density 0.018%

phrases related to suggesting or evaluating actions

phrases that express good ideas

No Comments

No Known Activations