INDEX

Explanations

proper nouns related to organizations and services

oai_token-act-pair · gpt-3.5-turbo

keywords related to health, consumer issues, and governance

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

ixtape

-0.72

 constitu

-0.67

 pronounced

-0.66

 coaster

-0.63

raints

-0.61

 judgement

-0.61

($)

-0.61

 vain

-0.60

 cruiser

-0.59

 redemption

-0.59

POSITIVE LOGITS

 Matters

0.77

Aid

0.76

Net

0.76

 Rooms

0.75

Week

0.72

Gear

0.71

 Services

0.71

Day

0.70

Hub

0.70

Bank

0.70

Activations Density 0.521%

proper nouns related to organizations and services

keywords related to health, consumer issues, and governance

No Comments

No Known Activations